Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 4 7 HAZARDOUS MATERIALS SHIPMENT INFORMATION 111 T. >-$ FOR EMERGENCY RESPONSE TRANSPORTATION RESEARCH BOARD National Research Council
1993 TRANSPORTATION RESEARCH BOARD EXECUTIVE COMMITFEE Chairman: A. RAY CHAMBERLAIN, Executive Director, Colorado Department of Transportation, Denver Vice Chairman: JOSEPH M. SUSSMAN, JR East Professor of Engineering, Massachusetts Institute of Technology, Cambridge Executive Director: THOMAS B. DEEN, Transportation Research Board MIKE ACOTT, President, National Asphalt Pavement Association, Lanham, Maryland (ex officio) ROY A. ALLEN, Vice President, Research and Test Department, Association of American Railroads, Washington, D.C. (ex officio) RICHARD E. BOWMAN, Acting Administrator, Maritime Administration, U.S. Department of Transportation (ex officio) E. DEAN CARLSON, Executive Director, Federal Highway Administration, U.S. Department of Transportation (ex: officio) JOSEPH M. DEL.BALZO, Acting Administrator, Federal Aviation Administration, U.S. Department of Transportation (ex officio) FRANCIS B. FRANCOIS, Executive Director, American Association of State Highway and Transportation Officials, Washington, D.C. (ex officio) JACK R. GILSTRAP, Executive Vice President, American Public Transit Association, Washington, D.C. (ex officio) THOMAS H. HANNA, President and CEO, American Automobile Manufacturers Association, Detroit, Michigan (ex officio) S. MARK LINDSEY, Acting Administrator, Federal Railroad Administration, U.S. Department of Transportation (ex officio) ROBERT H. MCMANUS, Acting Administrator, Federal Transit Administration, U.S. Department of Transportation (ex officio) ROSE A. MCMURRAY, Acting Administrator, Research and Special Programs Administration, U.S. Department of Transportation (ex officio) HOWARD M. SMOLKJN, Acting Administrator, National Highway Traffic Safety Administration, U.S. Department of Transportation (ex officio) LT. GEN. ARTHUR E. WILLIAMS, Chief of Engineers and Commander, U.S. Army Corps of Engineers, Washington, D.C. (ex officio) KIRK BROWN, Secretary, Illinois Department of Transportation, Springfield DAVID BURWELL, President, Rails-to-Trails Conservancy, Washington, D.C. L. G. (GARY) BYRD, Consultant, Alexandria, Virginia L. STANLEY CRANE, former Chairman and CEO of Consolidated Rail Corporation, Gladwyne, Pennsylvania RICHARD K. DAVIDSON, Chairman and CEO, Union Pacific Railroad, Omaha, Nebraska JAMES C. DELONG, Director of Aviation, Stapleton International Airport, Denver, Colorado JERRY L. DEPOY, Vice President, Properties and Facilities, USAir, Arlington, Virginia DON C. KELLY, Secretary and Commisioner of Highways, Transportation Cabinet, Frankfort, Kentucky ROBERT KOCHANOWSKI, Executive Director, Southwestern Pennsylvania Regional Planning Commission, Pittsburgh LESTER P. LAMM, President, Highway Users Federation, Washington, D.C. LILLIAN C. LIBURDI, Director, Port Department, The Port Authority of New York and New Jersey, New York City ADOLF D. MAY, JR., Professor and Vice Chair, Institute of Transportation Studies, University of California, Berkeley WILLIAM W. MILLAR, Executive Director, Port Authority of Allegheny County, Pittsburgh, Pennsylvania (Past Chairman, 1992) CHARLES P. O'LEARY, JR., Commissioner, New Hampshire Department of Transportation, Concord NEIL PETERSON, Executive Director, Los Angeles County Transportation Commission, Los Angeles DARREL RENSINK, Director, Iowa Department of Transportation, Ames DELLA M. ROY, Professor of Materials Science, Pennsylvania State University, University Park JOHN R. TABB, Director and CAO, Mississippi Department of Transportation, Jackson JAMES W. VAN LOBEN SELS, Director, California Department of Transportation, Sacramento C. MICHAEL WALTON, Paul D. and Betty Robertson Meek Centennial Professor and Chairman, Civil Engineering Department, University of Tcxas at Austin (Past Chairman, 1991) FRANKLIN E. WHITE, Commissioner, New York State Department of Transportation, Albany JULIAN WOLPERT, Henry G. Bryant Professor of Geography, Public Affairs and Urban Planning, Woodrow Wilson School of Public and International Affairs, Princeton University ROBERT A. YOUNG ifi, President, ABF Freight Systems, Inc., Fort Smith, Arkansas
Special Report 239 HAZARDOUS MATERIALS SHIPMENT INFORMATION FOR EMERGENCY RESPONSE Committee for the Assessment of a National Hazardous Materials Shipments Identification System Transportation Research Board National Research Council National Academy Press Washington, D.C. 1993
Transportation Research Board Special Report 239 Subscriber Categories I planning, administration, and environment IV operations and safety VII rail VIII freight transportation (multimodal) Transportation Research Board publications are available by ordering directly from TRB. They may also be obtained on a regular basis through organizational or individual affiliation with TRB; affiliates or library subscribers are eligible for substantial discounts. For further information, write to the Transportation Research Board, National Research Council, 2101 Constitution Avenue, N.W., Washington, D.C. 20418. Copyright 1993 by the National Academy of Sciences. All rights reserved. Printed in the United States of America NOTICE: The project that is the subject of this report was approved by the Governing Board of the National Research Council, whose members are drawn from the councils of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of the committee responsible for the report were chosen for their special competencies and with regard for appropriate balance. This report has been reviewed by a group other than the authors according to the procedures approved by a Report Review Committee consisting of the members of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. This study was sponsored by the Research and Special Programs Administration of the U.S. Department of Transportation. Library of Congress Cataloging-in-Publication Data National Research Council (U.S.). Committee for the Assessment of a National Hazardous Materials Shipments Identification System. Hazardous materials shipment information for emergency response / Committee for the Assessment of a National Hazardous Materials Shipments Identification System. p. cm. - (Special report ISSN 0360-859X; 239) ISBN 0-309-05421-4 1. Hazardous substancesâUnited StatesâTransportationâSafety measures. I. National Research Council (U.S.). Transportation Research Board. II. Title. III. Series: Special report (National Research Council (U.S.). Transportation Research Board) ; 239. T55.3.H3N29 1993 363.17âdc2O 93-15307 CIP Cover design: Tern Wayne Photograph credit, cover The Plain Dealer, Cleveland, Ohio
Committee for Assessment of National Hazardous Materials Shipments Identification System BARRY M. HoRowrrz, Chairman, MITRE Corporation, Bedford, Massachusetts MARK D. ABKOWITZ, Vanderbilt University, Nashville, Tennessee JOHN C. ALLEN, Battelle Memorial Institute, Cambridge, Massachusetts EDWARD R. CHAPMAN, Atchison, Topeka and Santa It Railway, Topeka, Kansas RANDOLPH M. Dol, Motorola Inc., Northbrook, Illinois LEE K. FooTE, Southern Chester County Development Foundation, Pennsylvania JOHN R. GUNTER, BellSouth Corporation, Atlanta, Georgia ROBERT J. HASKEN, General Electric Company, Fairfield, Connecticut DA!. L. HOWARD, Los Angeles City Fire Department, Los Angeles, California GERALD A. LONG, Roadway Express, Inc., Akron, Ohio JArvIES M. MELIUS, New York State Department of Health, Albany JOHN B. MORAN, Laborers' Health and Safety Fund of North America, Washington, D.C. SmPI-N B. STORMENTPhoenix Fire Department, Phoenix, Arizona DAN L. WILLIAMS, Off'ice of the State Fire Marshal, Springfield, Illinois Liaison Representatives Jol-IN D. FERRERA, Committee on Public Works and Transportation, U.S. House of Representatives RIcHA1W C. HANNON, Research and Special Programs Administration, U.S. Department of Transportation JOHN G. HART, U.S. Fire Administration, Federal Emergency Management Agency THoMAS HILLEGASS, Federal Highway Administration, U.S. Department of Transportation KEN HUNT, Occupational Safety and Health Administration, U.S. Department of Labor DONALD M. ITZKOFF, Committee on Commerce, Science, and Transportation, U.S. Senate TONY J0vER, U.S. Environmental Protection Agency Transportation Research Board Staff ROBERT E. SKINNER, JR. Director, Studies and Information Services JOSEPH R. MORRIS, Study Director THOMAS R. MENZIES, Program Officer NANCY A. ACKERMAN, Director, Reports and Editorial Services NORMAN SOLOMON, Assistant Editor
Preface I n the Hazardous Materials Trans-portation Uniform Safety Act of 1990, Congress called on the National Academy of Sciences "to con- duct a study of the feasibility and necessity of. . . a central reporting system and computerized telecommunications data center that is capa- ble of receiving, storing, and retrieving data concerning all daily ship- ments of hazardous materials, . . . and that can provide information to facilitate responses to accidents and incidents involving the transporta- tion of hazardous materials." The proposal that Congress asked the Academy to evaluate aims to improve information available to police and fire fighters at incidents involving release or threat of release of hazardous materials in trans- portation. It would create a national computerized central reporting system in which shippers and carriers would record information about each hazardous materials shipment as the shipment occurred. Police and fire fighters would have immediate access to this information when incidents took place. The system would complement or replace existing information sources, which include placards displayed on vehicles, documents carried by the vehicle operator, labels on packages, an emergency response manual prepared by the U.S. Department of Transportation (DOT), an industry-sponsored telephone referral ser- vice, and other resources. Reliable information about the presence and identity of hazardous materials is essential to emergency responders in V
viPreface making decisions about controlling hazardous materials transporta- tion incidents. The national central reporting system proposal was controversial. In statements to Congress and other public statements, proponents argued that the system would overcome inadequacies in existing pro- cedures for providing information that threaten the safety of fire fight- ers and the public. Opponents, including shipper and carrier industry groups and DOT, argued that the proposed system would be expensive to implement and vulnerable to the human errors that are the source of malfunctions in the existing system, that the existing information sys- tem works adequately, and that the information the national central reporting system would be intended to provide could affect the out- come of only a small fraction of hazardous materials transportation incidents. Hazardous materials whose transportation is subject to federal regu- lation include a wide variety of consumer products and industrial materials transported by rail, truck, water, or air in bulk and nonbulk quantities, the release of which could pose a threat to human health or safety. By weight, gasoline and other petroleum products account for most hazardous materials transported. Most hazardous materials transportation incidents to which police and fire fighters respond are of a minor nature, involving no injuries and limited property damage or other costs. However, incidents occur regularly with more severe conse- quencesâloss of life, injuries, and economic losses at least partially attributable to the presence of hazardous materials. Fire fighters are also concerned about possible long-term health effects of exposure to these materials. To conduct this study, the Transportation Research Board assembled a committee that included experts in hazardous materials safety, emer- gency response, freight transportation, and information systems. The committee considered various potential applications of communica- tions and information technology to aid emergency responders, strat- egies for evaluating and introducing information technology, and nontechnological options for improving information through better regulation, enforcement, and training. The committee recognized that a prerequisite to improving the exist- ing information system is an understanding of the information needs of emergency responders and the adequacy of current practices for meet- ing those needs. There has been little systematic evaluation of these subjects, and existing government programs for compiling data on the
Preface vii frequency and circumstances of incidents are not adequate to support evaluations. Therefore, in this study considerable effort was devoted to analyzing past hazardous materials transportation incidents. This study was conducted under the supervision of Robert E. Skin- ner, Jr., Director, Studies and Information Services, Transportation Research Board. Joseph R. Morris served as study director and drafted Chapters 1 and 4 and the Executive Summary under the guidance of the committee, which is responsible for all findings, conclusions, and rec- ommendations. Thomas R. Menzies conducted the analysis of haz- ardous materials transportation incidents and drafted Chapters 2 and 3. Committee members John Allen and Robert Hasken contributed to Chapter S. Funding for the study was provided by DOT as authorized in the Hazardous Materials Transportation Uniform Safety Act of 1990. The final report was prepared for publication under the supervi- sion of Nancy A. Ackerman, Director, TRB Reports and Editorial Services, and Norman Solomon, Assistant Editor, served as the editor of the report. Frances E. Holland and Marguerite E. Schneider typed the manuscript and provided other assistance throughout the study.
Contents Executive Summary Introduction 22 Study Charge, 23 History of the Reporting System Proposal, 25 Study Approach, 33 2 Hazardous Materials Transportation, Regulation, and Emergency Information 36 Characteristics of Hazardous Materials Transportation, 36 Hazardous Materials Regulation and Enforcement, 40 Emergency Response Information, 44 3 Necessity and Benefits of Improved Information 54 Data Sources, 55 Critical Information Needs of Responders, 63 Kinds of Information Problems Encountered by Responders, 66 Frequency of Information Problems, 79 Costs of Information Problems, 82 Summary, 95 4 Options To Improve Information: Technological Components 100 System Requirements, 101 Design Options, 103 Evaluation with Prototypes, 116
5 Options To Improve Information: Nontechnological Components 125 Regulatory Options, 126 Enforcement Options, 136 Training Options, 139 Summary, 150 Appendixes A Sections 1 and 25 of the Hazardous Materials Transportation Uniform Safety Act of 1990 151 B Data Needs and Sources 157 C National Transportation Safety Board Recommendations Pertaining to Emergency Response Information (1980 to 1992) 184 D Information Technologies in Transportation 191 E Shipper, Carrier, and Emergency Responder Information Capabilities 197 Study Committee Biographical Information 219
Executive Summary F ire fighters and police regularly respond to incidents involving the release or threat of release of hazardous materials in transport. Minimizing the danger of injury and other costs of such events requires knowledge of the materials so that appropriate fire fighting and other mitigation methods can be used and decisions made regarding evacua- tions or traffic diversion. Hazardous materials are substances that are flammable or toxic or have other properties that would threaten hu- mans if the substances were released. Government regulations require that certain shipments of hazardous materials be identified with placards on vehicles and shipping documents accompanying them, but lack of necessary information at the scene of an incident has sometimes interfered with response and led to increased losses. To improve information for responders at hazardous materials trans- portation incidents, creation of a national computerized central report- ing system and data center has been proposed. Shippers and carriers would be required by federal law to enter a record of each hazardous materials shipment, and police and fire fighters would have access to the information at the time of an incident. Congress called on the National Academy of Sciences to examine the feasibility and necessity of such a national central reporting system. The recommendations of the National Academy of Sciences commit- tee that conducted the study are listed below. All committee members
2 HAZARDOUS MATERIALS SHIPMENT INFORMATION are in agreement with each of the eight recommendations. However, members differed in their ranking of the relative importance of the recommendations and therefore in their preferences for the order in which the recommendations should appear. In particular, some mem- bers felt that Recommendation 8 should appear first because it is the recommendation most closely related to the specific question of the congressional charge for the study and because it is important to set boundaries on the kinds and scale of actions that are justifiable to correct identified problems. Other members believed that the presenta- tion should emphasize the need for improvements and recommenda- tions that lead to positive actions to improve hazardous materials safety (Recommendations 1-7). Because of these differences, the order of the recommendations should not be interpreted as reflecting a committee consensus on priorities. The committee believes that all its recommendations merit action and that acting on all of them would be the most efficient way to ensure progress in improving emergency responder access to vital information at hazardous materials transportation incidents. The recommendations are repeated with explanations later in this Executive Summary. Congress should provide the responsible agencies with the funding necessary to conduct the initiatives recommended. 1.Improvements to the existing system for providing information to emergency responders at hazardous materials transportation incidents are necessary. In a significant proportion of consequential incidents, responders are unable to obtain information they seek. Therefore, Congress, the U.S. Department of Transportation (DOT), and other responsible federal agencies should plan and carry out a program to improve the system. This program should include appropriate mea- sures to apply technology; reforms in regulation, enforcement, and training; and evaluation of the existing system so that efforts can be directed at the most pressing problems. DOT should immediately undertake one or more limited start- ups of automated information systems. The start-ups should be con- trolled experiments with rigorously designed evaluation protocols. A practical limited start-up is identified in the last section, involving an information system for rail shipments (see p. 15). DOT should, on an ongoing basis and in conjunction with the shipper and carrier industries and emergency responders, system- atically investigate opportunities for application of information tech-
Executive Summary 3 nology to aid emergency responders and reduce the costs of hazardous materials incidents. 4. The U.S. Fire Administration of the Federal Emergency Manage- ment Agency, DOT, and other federal, state, and local agencies that maintain data bases of hazardous materials incidents should formally coordinate to ensure that data are defined and collected uniformly, duplicate reporting is avoided, and data collection is designed to serve essential program evaluation and research needs. S. DOT should establish a monitoring capability that allows it to determine whether its regulations intended to provide emergency re- sponders with information at hazardous materials transportation inci- dents are working adequately. Monitoring should be planned and conducted with state and local government participation. DOT, together with other responsible federal agencies, should form a plan of action to alleviate each of the six categories of informa- tion failures identified in this study (see p. 7) through changes in regulations, more effective enforcement, and support for improved training of emergency responders and inspectors. Possible components of this action plan have been identified (see pp. 18-20). DOT and the U.S. Fire Administration, with the full participation of state and local emergency response agencies, should jointly conduct a study of costs and means of organizing and delivering training to hazardous materials emergency responders and enforcement officers. The government should not attempt to implement a system such as the originally proposed national central reporting system, that is, one entailing immediate and universal application of a requirement for shipper or carrier real-time filing of vehicle contents information in a central data base. STUDY APPROACH AND SCOPE The approach to the study was to analyze The value of improved information by identifying (a) the informa- tion that emergency responders need to make decisions at incidents; the performance of existing means of providing information; and the frequency, causes, and consequences of information failures; Technological options that might offer solutions to information problems, including the national central reporting system and other
4 HAZARDOUS MATERIALS SHIPMENT INFORMATION designs, by considering the cost, technical feasibility, and organiza- tional requirements for implementing such systems; and o Regulatory, enforcement, and training options for solving the information problems, that is, refinements within the existing system for identifying hazardous materials that would not necessarily depend on adoption of information technology. The scope of the study was limited to matters that may affect the consequences of hazardous materials incidents after they occur, and not to methods of preventing incidents. Incident prevention is funda- mental to reducing risks to emergency responders and the public. The methods examined in this report for improving emergency response in general would not directly contribute to or substitute for accident prevention programs. Also, the scope was limited to the safety of materials in transport rather than to safety of materials in use or storage at fixed facilities or in storage in transport vehicles or con- tainers. Although the committee did not focus exclusively on any particular system design, it did focus on prospective systems that would be national in application. Programs aimed at incident prevention and safety of fixed facilities also require supporting information systems. For example, incident prevention depends on analysis of the causes of past incidents, which requires a historical incident data base. For fixed facilities, local govern- ments are already compiling data bases of hazardous materials stored in their jurisdictions, using information provided by hazardous mate- rials users in accordance with Environmental Protection Agency (EPA) regulations. Also, information systems to record exposures of emer- gency responders to toxic materials might have value for studies and monitoring of responder health. Information systems for incident pre- vention, fixed facilities, monitoring of exposure, and response to trans- portation incidents might eventually be most effective as components of a single system. This study did not consider such comprehensive data needs or estimate costs and benefits of such a comprehensive informa- tion system. The scope of the investigation of information needs and failures was limited further to public-sector emergency responders, as opposed to private responders (e.g., transportation company employees who deal with incidents occurring on company property), because assistance to public-sector responders was the objective of the original central re-
Executive Summary S porting system proposal and public-sector requirements would dictate the design of any new information systems. The conclusions are relevant to all modes of freight transportationâ air and water as well as truck and rail. However, since nearly all hazardous materials transportation incidents involve truck or rail, data collection and analysis concentrated on these modes. NECESSITY AND VALUE OF IMPROVED INFORMATION There has never been a comprehensive evaluation of the information needs of emergency responders at hazardous materials transportation incidents or of the performance of the existing system in providing information. Before an improved system can be designed, it will be necessary to understand responders' information requirements at inci- dent scenes and why current practice (relying on placards displayed on vehicles, labels on packages, documents carried in vehicles that identify hazardous materials on board and give emergency response informa- tion, and a mandatory 24-hr emergency telephone number for access to the shipper) sometimes fails to meet those requirements. To begin to fill these gaps in understanding, incident data bases were reviewed, emergency responders and managers in industries responsi- ble for hazardous materials handling were interviewed, and case studies of consequential incidents (i.e., those involving injury, substantial property damage, evacuation, or major traffic delay) were conducted. The case studies were supplemented with descriptions of consequential incidents investigated by the National Transportation Safety Board (NTSB). The total of 125 cases are representative examples of all incidents of comparable severity. The case study data have important limitations: the size of the sample was small considering the diversity of incident circumstances; NTSB does not randomly select the cases it investigates; information about incidents and consequences was often incomplete; follow-up conducted for this study was limited and was conducted some time after the incidents occurred; and any long-term or delayed health effects of exposures to materials would have been especially difficult to observe. Therefore, it is likely that some consequences were over- looked, and all conclusions drawn from the data must be regarded as tentative. Nonetheless, the data and analysis are more complete than in any previous examination of the issues.
HAZARDOUS MATERIALS SHIPMENT INFORMATION Hazardous Materials Transportation Safety Roughly 10,000 to 20,000 truck transportation incidents and roughly 1,000 to 1,500 rail incidents occur each year that involve release of a hazardous material or a circumstance that threatens a release and to which public-sector emergency responders are dispatched. In the ma- jority of these incidents, the costs are confined to cleanup and use of emergency response resources. As many as 1,000 incidents annually have greater consequences: injury, substantial property damage, evac- uation, or major traffic delay (only part of these costs is attributable to the presence of the hazardous material). In the case studies, there was an average of two reported injuries per consequential incident. In most consequential incidents no injuries were reported, but in a few, there were numerous injuries. About half the injuries in the case studies appeared to be related to the presence of the hazardous material. Information Requirements Information needs are driven by the decisions that emergency re- sponders must make: what protective equipment is needed to approach the incident; whether danger of toxic exposure, fire, or explosion justifies a highway closing or evacuation; whether to use water or other chemicals to extinguish a fire; how to stop and contain a release; and how to dispose of spilled material. The case studies show that responders need to determine the follow- ing critical information items: whether hazardous materials are present in the incident, the general type of hazard (e.g., flammable, poisonous), and the specific identities of the hazardous materials. Responders may also need other facts about the shipment, information about the sur- roundings, and expert advice. These needs can be divided into two categories: information specific to the vehicles involved (e.g., identity of materials) and information that is not vehicle specific (e.g., properties of materials). Provision of vehicle-specific information is the defining charac- teristic of the original national central reporting system proposal. Performance of the Existing System The case studies indicate that the existing system fails with some regularity, and these failures can have costs. Perhaps most significant,
Executive Summary 7 many emergency responders interviewed expressed lack of full confi- dence in the system. Among the case studies, in about 25 percent of consequential high- way incidents and 10 percent of consequential rail incidents, emer- gency responders were unable to obtain information that they sought or experienced significant delay in obtaining it. In the remaining inci- dents, the existing system performed adequately: responders obtained credible information in the format required by regulations. In those incidents with information deficiencies, the adverse consequences were not necessarily caused by the deficiencies. With the percentage of deficiencies just cited, roughly 200 consequential incidents would oc- cur annually nationwide in which emergency response information would be seriously deficient. Six categories of information failures appeared in the case studies: Required sources of information, such as placards and shipping papers, were missing or inaccurate. Placards, shipping papers, or other information sources were ob- scured, destroyed, or inaccessible because of a crash or fire or threat of fire, explosion, or toxic exposure. Information sources were in compliance with regulations and accessible, but they failed to efficiently convey important information to responders. Information was insufficient because the material or shipment was exempt from some federal hazardous materials transportation regula- tions (because quantities were too small to require placarding, the material was not regarded as being in transport, or the material was not designated as hazardous in regulations). The vehicle operator was unprepared to provide information. Responders failed to obtain or use available information. Because of uncertainty in assessing the consequences of failures in the case studies and other limitations of the data, ranking the severity of these problems is difficult. However, the most common problems in the case studies were the first, third, and fourth categories listed above: required information was missing, information was in compliance and accessible but was not adequate for responders' needs, or information was insufficient because the shipment was exempt from some regula- tions. Instances of destroyed or inaccessible papers and placards were
HAZARDOUS MATERIALS SHIPMENT INFORMATION much less common because the great majority of incidents do not involve a serious crash or fire. Results of roadside inspections to enforce truck regulations support the case study finding that regulatory violations are a common problem. In inspections, roughly one-third of trucks required to be placarded are cited for some violation of material identification regulations. The frequency of information problems is not uniform among all segments of transportation. For example, a common kind of incident with information problems involves trucks carrying mixed loads. Mixed loads present problems because placards do not identify specific hazards when quantities are small, shipping papers can be voluminous when a truck is carrying multiple shipments, drivers are less likely to be familiar with cargoes, the appearance of the truck does not give infor- mation about contents, and responders must be concerned with haz- ards from interactions of materials. Because quantities are smaller, mixed-load incidents tend to be less severe than bulk-load incidents, but response is often protracted because of information problems. Benefits of Improved Information Improving the reliability of hazardous materials identification could reduce consequences in extreme circumstances by helping to ensure that actions are taken to protect the public and fire fighters from threat of injury. Because of the limitations of small sample size and incomplete data, the case study analysis did not provide a firm basis for estimating the linkage between information and injuries and illnesses. In the case study incidents, most reported injuries were apparently unrelated to the presence of the hazardous materials (e.g., injuries from impact in motor vehicle crashes) or occurred nearly simultaneously with the onset of the incident as the result of explosion or fire and before any emergency response could have occurred. However, some injuries re- lated to the presence of the hazardous material occurred after some delay and thus might conceivably have been avoided through improved emergency response effectiveness. Hazardous materials incidents have been documented where faulty emergency response information led to injuries or deaths, although such circumstances were not evident among the 125 case studies specially developed for this study. The case studies suggest that the greatest opportunity for avoiding injuries through improved information is through more reliable initial warning
Executive Summary 9 of the presence of the hazardous materialâthe function that the placard is intended to perform. In more routine situations, more reliable information could reduce the need for evacuations or traffic diversions and allow more efficient use of emergency response resources. To reduce costs, an improved information system would have to produce an increase in reliability great enough to change the actions of emergency responders. The incident case studies indicate that emergency responders tend to take a cautious course of action, in part because they believe from experience that materials information can be unreliable. Responder actions may include precautions that have significant costs (e.g., traffic blockages or workplace evacuations) but are believed to be warranted to reduce the risk of injury or death as much as possible. Because of the cautious practices of emergency responders, and because responders often do not have a wide range of options for dealing with an incident, small improvements in reliability or com- pleteness of information might not alter responders' actions in many incidents. On the other hand, if the improvement in reliability were great enough to boost overall responder confidence, the result might be more efficient responseânot just to those incidents in which informa- tion failures would otherwise occur, but in other instances where responders now take precautions that have costs because they do not trust available information. For the estimated 200 annual consequential incidents with informa- tion problems, the combined costs of property damage, evacuations, traffic delays, productivity losses, and response personnel and equip- ment amount to several hundred million dollars. Data are insufficient to allow the estimation of environmental damage and cleanup costs. In addition, these incidents are associated with roughly 400 reported acute injuries and illnesses. Some of these costs are unrelated to the hazardous nature of the cargoes. The case study incidents suggest that improved information could prevent a fraction of the dollar costs of consequential incidents with information problems, for an annual sav- ings in the tens of millions of dollars. Also, improved information probably would lead to more efficient use of emergency response resources in all hazardous materials transportation incidents, for sav- ings that might amount to some additional millions of dollars annually. Finally, some fraction of the injuries and illnesses attributable to these incidents might be avoided through better information, but the data do not allow an estimate of this fraction.
10 HAZARDOUS MATERIALS SHIPMENT INFORMATION These estimates, derived from the case studies, do not include possi- ble benefits from reduced consequences of rare but catastrophic events. In the United States from 1946 to 1989 there were at least five truck or rail incidents in each of which 15 or more people were killed as a result of the presence of a hazardous material. Other severe incidents have occurred in waterborne shipping. A majority of these fatalities and incidents involved explosions of flammable liquids or gases nearly simultaneously with the onset of the incident, and improved emergency responder information would be unlikely to prevent deaths in such circumstances. However, greater information reliability and availabil- ity might help avoid some consequences of such catastrophes. A 1988 incident in which six fire fighters were killed is one in which improved information might have saved lives. This incident was beyond the scope of the originally proposed national central reporting system because the hazardous materials involved were not being transported, but were stored in trailers and so were not subject to DOT hazardous materials transportation regulations. Therefore, to be useful in this incident, an information system would have had to cover hazardous materials in vehicles used for storage as well as materials in transporta- tion. The deaths were caused by an explosion of blasting agents stored in two trailers set afire by an arsonist at a construction site. The trailer did not bear DOT placards, and in fact standards in effect at the time called for the removal of placards on explosives stored in vehicles subsequent to transportation. An investigation concluded that the fire fighters were informed en route of the possibility of explosives and apparently learned on the scene that the burning trailer was supposed to contain explosives; however, the investigation concluded that before the alarm, the fire department did not have knowledge of the type, quantities, or locations of the blasting agents. The availability of such information at the time of the alarm might have allowed fire fighters to plan how to deal safely with the incident. This incident also illustrates the need for improved coordination among DOT and the agencies that regulate identification of hazardous materials not in transport. Efficacy of Possible Solutions The primary intent of the original proposal for a national central reporting system was to allow material identification when papers or placards were inaccessible or destroyed, provided that the vehicle could
Executive Summary 11 be identified and the responder knew or suspected that hazardous materials were present. The problem of inaccessible or destroyed pa- pers or placards is one of the six categories of information failure identified in the case studies. The other five categories of information failure are related to underlying problems in regulations, enforcement, or training: regulatory requirements governing shipment identification do not match emergency responders' information needs; regulations are violated; or inadequate training of shipper, carrier, or emergency response personnel results in inappropriate actions. The success of the existing system in most incidents indicates that regulatory refinements and strengthened enforcement and training would improve perfor- mance. Development of a technology-based information system would not be a substitute for needed regulatory, enforcement, or training im- provements, and problems in the existing system arising from such sources as clerical errors or negligence could be replicated in any new system. However, technological aids might complement regulatory, enforcement, and training improvements. For example, an information system might help responders deal with complex situations such as mixed loads by automatically organizing information to highlight per- tinent facts, it might strengthen enforcement by creating a monitoring mechanism, or it might have features designed to reduce the likelihood of human errors. Of course, the introduction of new systems would create new regulatory, enforcement, and training needs. FEASIBILITY AND COST OF TECHNOLOGY APPLICATIONS Attempts to introduce automation into an established system such as the existing hazardous materials emergency response information sys- tem are unlikely to be successful unless three guidelines are followed. First, the advantages of automation are achieved only when the enterprise to be automated is well understood and well managed. If there is confusion about the process, efforts to automate often result in greater confusion. Performance criteria for an improved system can be defined only by studying the performance of the existing system. There- fore, in the near term, efforts must be made to rationalize the existing information system and to measure its performance.
12 HAZARDOUS MATERIALS SHIPMENT INFORMATION Second, the evolution of an information system should, as a primary factor of design, identify and take advantage of opportunities created by investments in other technological developments. Building on exist- ing information systems in the transportation industry would be the most practical approach to a hazardous materials information system. Making use of systems that were in routine daily use for business purposes would be the best way to attain reliability, and the costs of such a strategy would be much less than the costs of procuring a stand- alone system. Third, automation should be introduced in an existing information system in carefully planned steps. This approach allows for modifica- tions to resolve issues of human error, diligence, and training on the basis of evaluations of actual experience. Initial steps could be made with those segments of the community who have already invested in automated information systems and are experienced in their use. This stepwise approach would provide a basis of experience for making cost- benefit trade-offs. To be most cost-effective, initial steps would ideally also target segments in which the potential payoff from improved information was greatest. The originally proposed national central reporting system would not conform with any of these guidelines. Because knowledge about the information needs of emergency responders and the causes of failures of the existing system is inadequate, there is no basis for determining a worthwhile increment in reliability or establishing other performance criteria for the new system. The originally proposed system would not be a phased implementation, and it would duplicate rather than inte- grate existing shipper and carrier information capabilities. Building on existing capabilities is critical because of the rapid pace of introduction of innovative information technology in transporta- tion. Some firms in certain segments of the carrier and shipper indus- tries could comply today with most of the record-keeping and reporting requirements necessary to participate in a modified national central reporting system or in the clearinghouse option described in this report at a cost that would not be deeply disruptive. Throughout the transpor- tation industry, carriers and shippers are rapidly developing new capa- bilities for maintaining current records on goods in transit to reduce inventory costs, for monitoring vehicles to allow more efficient fleet utilization, and for eliminating paper transactions by use of electronic data interchange.
Executive Summary 13 These capabilities differ greatly by size of firm, industry, and mode. Rail is probably the most advanced, because automated car-tracking systems have been widely implemented. In other sectors, many firms have little or no shipment-tracking capability, and imposition of a national central reporting system requirement would force them to greatly alter their practices, increasing their costs. Because many firms lack the information systems that would be required for compliance with a national central reporting system, rapid full-scale implementation of the system, which was a key feature of the original proposal, would be very unlikely to succeed. Firms would require time to develop, test, and perfect the needed internal informa- tion systems. The cost components of an automated hazardous materials ship- ment information system would be shipper and carrier internal infor- mation systems, the central service facility, the emergency responder interface, enforcement, and training. The cost of shipper and carrier internal systems would be the largest component, much greater than the cost of the central facility. Therefore, the net cost would be highly sensitive to the extent to which shippers and carriers could use their existing investments in information technology to comply with system requirements. The cost also would depend on the functions the system supported; the number of shippers, carriers, and responders participat- ing; the categories of shipments covered; and the level of reliability required. Cost estimates can only be illustrative and based on arbitrary assumptions, because the appropriate scale and functioning of the system could only be determined in an evolutionary process starting with test implementations. Estimates were made for a hypothetical system including only larger emergency responsc agencies and shippers and carriers handling haz- ardous materials with some regularity. This would be a much more modest system than that envisioned in the original national central reporting system proposal. If this hypothetical system were adopted today, initial capital and start-up costs would be on the order of $300 million and annual operation costs would be $250 million. If such an expenditure were made immediately, without thorough preliminary testing and evalua- tion, there would be no assurance that the system would produce a meaningful improvement in reliability or that its value would justify the cost.
14 HAZARDOUS MATERIALS SHIPMENT INFORMATION Most of the estimated costs are for shipper and carrier information systems. However, the information system that shippers and carriers would have to install to comply with automated hazardous materials tracking requirements would be similar in some respects to systems that a few trucking companies and nearly all railroads have already adopted to meet internal management and customer service needs. Although such systems will not be ubiquitous in transportation within the foreseeable future, their use is spreading rapidly. Carriers with such capabilities already in place probably could avoid most of the costs of compliance that would be incurred by comparable carriers with no such capabilities. RECOMMENDATIONS Actions Needed To Improve Hazardous Materials Emergency Response Information Improvements to the existing system for providing information to emergency responders at hazardous materials transportation incidents are necessary. Therefore, Congress, DOT, and other responsible federal agencies should plan and carry out a program to improve the system. This program should include appropriate measures to apply technology; reforms in regulation, enforcement, and training; and evaluation of the existing system so that efforts can be directed at the most pressing problems. Although this approach will take sometime, it is the quickest means to bring about improvement, since attempts to implement improvements without analysis of needs and without test- ing would probably fail. Several of the following recommendations call for data collection and evaluations. The committee has not estimated the costs of all of these activities in detail. However, their costs would be justified by the value of the reduction in the risks of hazardous materials trans- portation that could stem from them. As a first step, DOT and the other responsible agencies should estimate the costs of the recom- mended activities and submit proposals to Congress for funding. Congress should provide the necessary funds specifically for these purposes and should provide direction to the agencies responsible for carrying out the recommendations.
Executive Summary 15 Implementing New Information Systems DOT should immediately undertake one or more limited start-ups of automated information systems. Limited start-ups should be con- trolled experiments with independent, rigorously designed evaluation protocols. Their design, conduct, and evaluation should be governed by cooperative arrangements among participating federal, state, and local government agencies and participating shippers and carriers. Each limited start-up should be planned as the foundation for a widely implemented system, provided it is demonstrated to have benefits sufficient to justify its costs. Examples of possible limited start-ups are Emergency responder access to carrier shipment-tracking systems via an established information network service; Incorporation of hazardous materials monitoring functions in truck automatic electronic identification demonstration projects, which federal and state transportation agencies now have under way or planned; and Emergency responder access to rail industry car-tracking and way- bill data systems. The rail industry limited start-up would be a practical starting point. All large railroads provide current information to shippers describing the location and recent movements of railcars carrying their shipments. The information can be provided over computer networks or via syn- thesized voice messages over ordinary telephones. The system could be modified to provide information on car contents to emergency responders. A limited start-up of a rail system would be relatively straightfor- ward to organize. It would demonstrate a service that emergency responders may find valuable and would provide an experimental setting for addressing many of the organizational, technical, and cost- effectiveness issues that any emergency response information system would face. The other two suggested limited start-ups would be feasible, and in fact testing of hazardous materials applications is likely to be a compo- nent of the current truck automatic electronic identification demon- strations. However, the infrastructure and parallel industry investment for expanding these two beyond the prototype stage are only beginning
16 HAZARDOUS MATERIALS SHIPMENT INFORMATION to develop, so they would not progress beyond the experimental stage for some time. Also, a limited start-up using an established information network service in trucking probably would require more investment in developing new systems than would the rail limited start-up, because the services now available fall short of the requirements of a hazardous materials information system. The limited start-up should have federal government, industry, and emergency responder participation. Key participants would be at least one large railroad and several emergency response agencies in the railroad's territory. All participants, including responders, should be involved in defining performance criteria and planning the evaluations. Evaluation should include comparison of the timeliness and reliability of materials identification via the limited start-up system with those of current methods. The limited start-up could also compare alternative communications links between responders and the rail data base (in- cluding human interaction), alternative information resources for in- terpreting the cargo information received, and arrangements for data security. The cost of a 2-year limited start-up involving one railroad, including costs to the railroad and emergency responders and costs for evalua- tion, is estimated to be $4 million. If evaluation of this trial showed evidence of reduction in incident costs, the limited start-up could be expanded in the third year to a trial of a nationwide system involving all major railroads. The second phase would bring the total cost to roughly $7 million. DOT should, on an ongoing basis and in conjunction with the shipper and carrier industries and emergency responders, system- atically investigate opportunities for application of information tech- nology to aid emergency responders and reduce the costs of hazardous materials incidents. The guidelines described above should be applied to any technology application: performance criteria concerning the information to be provided and the reliability required must be defined by studying the performance of the existing system; automation should be introduced in planned steps to allow for evaluation and modifica- tions to resolve problems; and applications should take advantage of developments in information and communication systems used by shippers, carriers, and responders. A further critical criterion in system design is provision for security of proprietary business data. DOT should track technical developments in transportation information and
Executive Summary 17 communication systems to ensure that opportunities for applications to improve hazardous materials safety are not overlooked. Data and Evaluation The U.S. Fire Administration, DOT, and the other federal, state, and local agencies that maintain data bases of hazardous materials inci- dents should formally coordinate to ensure that data are defined and collected uniformly, duplicate reporting is avoided, and data collection is designed to serve essential program evaluation and research needs. In keeping with its current involvement in programs to assemble nation- wide incident data, the U.S. Fire Administration would be the appropri- ate agency to lead this coordination. DOT and Fire Administration data systems should be made compatible so that they can be used jointly in analyses, and new efforts should be undertaken to ensure reliability of data collected in the existing systems. DOT should establish a monitoring capability that allows it to determine whether its regulations intended to provide emergency re- sponders with information at hazardous materials transportation incidents are working adequately. DOT needs to understand the infor- mation requirements of emergency responders, how frequently and why existing practices fail to provide that information, and the costs of information failures. Developing this understanding is a prerequisite to reducing the costs of hazardous materials incidents by improving infor- mation available to responders. DOT's Hazardous Materials Information System (HMIS) was de- signed for purposes other than this monitoring function. For a variety of reasons having to do with reporting requirements and compliance, HMIS does not record the majority of hazardous materials incidents that involve public-sector emergency response and thus cannot serve as the monitoring system. Therefore a new data collection method is needed. It should be founded on a statistically valid sampling pro- cedure. Rather than attempt to maintain a census of hazardous mate- rials incidents for this purpose, a more practical approach could be for DOT to select a sample of emergency response agencies and investigate a sample of incidents chosen from the records of each agency. DOT should plan its monitoring of the effectiveness of the existing system to permit comparisons with the evaluations of the recommended limited start-up system. The monitoring should be planned and conducted
18 HAZARDOUS MATERIALS SHIPMENT INFORMATION with advice from and participation of state and local emergency responders. Regulation, Enforcement, and Training DOT, together with the other responsible federal agencies, should form a plan of action to alleviate each of the six categories of information failures identified in this study through changes in regulations, more effective enforcement, and sup port for improved training of emergency responders and inspectors. Effective regulations and adequate training and enforcement will be essential components of any improvement in emergency response effectiveness. It was not possible in this study to assess the costs and benefits of specific possible regulatory, enforce- ment, and training actions. However, DOT should evaluate the pro- posals in the following table. The evaluations should estimate the costs to all parties of carrying out the regulatory, enforcement, and training actions and the potential benefits in terms of improved information and resulting reductions in the harmful effects of hazardous materials incidents. If DOT finds that the proposals are inadequate, it should devise other means that would correct the problems and yield benefits justified by the costs. Problem Proposal Required sources of in- Systematically monitor compliance as the basis formation are missing for establishing targets for compliance im- or inaccurate provement and measuring progress toward them Upgrade training of motor vehide inspectors to en- sure more uniform enforcement Improve effectiveness of enforcement of haz- ardous materials identification requirements in vehicle safety inspections Monitor the effectiveness of required shipper and carrier employee training programs Establish higher fines for certain hazardous ma- terials identification violations, especially im- proper placarding and failure to provide tele- phone contact Facilitate the filing of complaints by emergency responders who discover violations at incidents
Executive Summary 19 Problem Proposal Placards, shipping pa- Improve placard visibility through the use of ret-' pers, or other infor- roreflective materials and other means marion sources are Evaluate new materials for prolonging the fire re- obscured, destroyed, sistance of placards or inaccessible be- - cause of a crash or fire or threat of fire, explosion, or toxic ex- posure Information sources are Ensure that shipping papers and placards on ve- in compliance with hides carrying mixed loads are comprehensible regulations and acces- to emergency responders by (a) requiring a sible, but they fail to summary manifest listing critical information efficiently convey about each hazardous shipment on board a ye- important information hide, (b) requiring carriers to provide 24-hr to responders telephone contact to a person with access to vehicle cargo information, (c) replacing the Dangerous placard with a more specific mes- sage, and (d) changing the minimum quantities requiring placarding Analyze the use of CHEMTREC and the National Response Center by local emergency responders to determine whether needed services are being provided and develop more effective referral services Coordinate material identification requirements among DOT, EPA, and the Occupational Safety and Health Administration to ensure that they are consistent and understood by re- sponders; in particular, eliminate ambiguities concerning identification of hazardous contents of trailers or railcars being used as storage Develop a standardized material data sheet and require that it accompany shipping papers for certain hazardous materials Information is insuffi- - Encourage industry efforts to adopt uniform cient because the ma- tank trailer numbering terial or shipment is exempt from some federal hazardous ma- terials transportation regulations
20 HAZARDOUS MATERIALS SHIPMENT INFORMATION Problem Proposal Vehicle operator is un- Ensure that training of shipper and carrier per- prepared to provide sonnel covers communication with local emer- information gency responders Responders fail to ob- Expand federal resources and technical assistance tain or use available for training of responders in the skills of haz- information ard identification and interpretation Develop a hazardous materials identification training curriculum DOT and the U.S. Fire Administration should jointly conduct a study of costs and means of organizing and delivering training to hazardous materials emergency responders and enforcement officers. Participation of state and local emergency responders in the study must be ensured. Also, participation of EPA and the National Institute of Environmental Health Sciences would be valuable. Training will be essential to the success of any attempts to improve the emergency response information system, including those that would use automa- tion. Lack of training is a factor in most of the information problems that arise in hazardous materials incidents. The study should be under- taken with the oversight of Congress, and its product should be a report to Congress on national requirements and the federal role in training. Original National Central Reporting System Proposal The government should not attempt to implement a system such as the originally proposed national central reporting system, that is, one entailing immediate and universal application of a requirement for shipper or carrier real-time filing of vehicle contents information in a central data base. There is no sound basis for defining performance criteria for information to be provided and threshold reliability needed in such a system; there would be no opportunity to allow ongoing evaluation to guide implementation, as a phased implementation would allow; and the system would not be designed to make maximum use of existing shipper, carrier, and responder capabilities. Also, the original national central reporting system proposal is not aimed at the most serious failures of the existing system. The incident case study results tentatively indicate that the failure the proposal was primarily intended to correctâincidents in which papers or placards are mac-
Executive Summary 21 cessible because of a crash or fireâis only one category of failure, and not the most frequent category, under the existing system. Because of these shortcomings, the originally proposed system would be unlikely to function as intended or to produce benefits sufficient to justify its costs. Information technology can make important contribu- tions to emergency response, but this.potential is most likely to be attained through the phased approach recommended earlier.
Introduction F ire fighters and police often must respond to accidents involving vehicles carrying hazardous materials or to other incidents involving the release or threat of release of hazardous materials in transportation. Dealing with such an event requires knowledge of the nature of the materials so that appropriate fire fighting, cleanup, and disposal methods can be used; decisions about evacuations or traffic diversion made; and danger of injury, death, or property damage minimized. Regulations require that certain shipments of hazardous materials be identified by placards on vehicles and shipping documents accompany- ing them, but lack of necessary information at the scene of incidents has sometimes delayed response and increased risks. To improve information at hazardous materials transportation inci- dents, a proposal has been made for creation of a national comput- erized central reporting system and data center, in which shippers and carriers would be required by federal law to enter a record of each hazardous materials shipment at the times of commencement and completion of the shipment. The proposal would not require tracking the geographic position of vehicles. At the time of an incident, local police and fire fighters would have immediate access to the data center. Through the vehicle's license plate or other number, they could identify any hazardous materials carried by the vehicle or learn other vehicle- 22
Introduction23 specific information. The proposal has been introduced in legislation in the U.S. Congress, although not enacted. The following section describes the congressional charge to the com- mittee. The next section describes the origin and history of the national central reporting system proposal. The final section ' describes the ap- proach that was taken in conducting this study of hazardous materials transportation information systems. STUDY CHARGE In the Hazardous Materials Transportation Uniform Safety Act of 1990 [Public Law 101-615 (Nov. 16, 1990)], Congress directed the Secretary of Transportation to undertake to enter into appropriate arrangements with the National Academy of Sciences to conduct a study of the feasibility and necessity of establishing and operating a central reporting system and comput- erized telecommunications data center that is capable of receiving, storing, and retrieving data concerning all daily shipments of haz- ardous materials, that can identify hazardous materials being trans- ported by any mode of transportation, and that can provide information to facilitate responses to accidents and incidents involving the transporta- tion of hazardous materials. The act asks that the relevant federal regulatory agencies, affected industries, and state and local law enforcement and fire fighting organi- zations be consulted in the conduct of the study and that a report be submitted including "recommendations of the National Academy of Sciences with respect to establishment and operation of a central reporting system and computerized telecommunications data center." The act includes a list of specific questions concerning technical and administrative issues in the operation of the system for the study to address (see accompanying text box and Appendix A). The act directs the Secretary of Transportation to use the study report in a rule making. The rule making is not to lead to a rule on implementing a tracking system, but rather to a recommendation from the secretary to Congress on whether a system should be implemented.
Public Law 101-615, 101st Congress, November 16, 1990 (104 Stat. 3244) SEC. 25. IMPROVEMENTS TO HAZARDOUS MATERIALS IDENTIFICATION SYSTEMS CENTRAl. REPORTING SYSTEM AND COMPUTERIZED TELECOMMUNICATIONS DATA CENTER STUDY.â (1) ARRANGEMENTs WITH NATIONAL ACADEMY OF SCIENCES.âNot later than 30 days after the date of the enactment of this Act, the Secretary of Transportation shall undertake to enter into appropriate arrangements with the National Academy of Sciences to conduct a study of the feasibility and necessity of establishing and operating a central reporting system and computerized telecommunications data center that is capable of receiving, storing, and retrieving data concerning all daily shipments of hazardous materials, that can identify hazardous materials being transported by any mode of transportation, and that can provide information to facilitate responses to accidents and incidents involving the transportation of hazardous materials. ADDITIONAL PURPOSES OF RULEMAIUNG PROCEEDING AND STIJDY.âAdditional pur- poses of the rulemaking proceeding initiated under subsection (a) with respect to a central reporting system and computerized telecommunications data center described in subsec- tion (b) and the study conducted under subsection (b) areâ to determine whether such a system and center should be established and operated by the United States Government or by a private entity, either on its own initiative or under contract with the United States; to determine, on an annualized basis, the estimated cost for establishing, operat- ing, and maintaining such a system and center and for carrier and shipper compliance with such a system; to determine methods for financing the cost of establishing, operating, and maintaining such a system and center; to determine projected safety benefits of establishing and operating such a system and center; to determine whether or not shippers, carriers, and handlers of hazardous materials, in addition to law enforcement officials and persons responsible for respond- ing to emergencies involving hazardous materials, should have access to such system for obtaining information concerning shipments of hazardous materials and technical and other information and advice with respect to such emergencies; to determine methods for ensuring the security of the information and data stored in such a system; to determine types of hazardous materials and types of shipments for which information and data should be stored in such a system; to determine the degree of liability of the operator of such a system and center for providing incorrect, false, or misleading information; to determine deadlines by which shippers, carriers, and handlers of hazardous materials should be required to submit information to the operator of such a system and center and minimum standards relating to the form and contents of such information; to determine measures (including the imposition of civil and criminal penalties) for ensuring compliance with the deadlines and standards referred to in paragraph (9); and to determine methods for accessing such a system through mobile satellite service or other technologies having the capability to provide 2-way voice, data, or facsimile services.
Introduction 25 HISTORY OF THE REPORTING SYSTEM PROPOSAL The proposal for a national central reporting system has been discussed for more than 15 years. A description of the history of the proposal is valuable in understanding why Congress saw a need for a study by the National Academy of Sciences. The Hazardous Materials Transportation Act of 1975 (49 USC 1801) forms the basis for current federal regulation. It consolidated responsibilities previously dispersed among numerous agencies under the U.S. Department of Transportation (DOT); provided for definition of regulated materials; authorized regulation of labeling, placarding, packages and containers, and handling practices; provided enforce- ment powers; and preempted inconsistent state rules. The act also called for a "central reporting system and data center" to provide "law enforcement and fire fighting personnel of commu- nities, and other interested persons and government officers, with technical and other information and advice for meeting emergencies connected with the transportation of hazardous materials" (Haz- ardous Materials Transportation Act, Section 109). A similar provision had been enacted in 1970 (U.S. Congress. Senate. 1979, 112). This language appears to require some direct service to emergency re- sponders, and presumably on-the-scene services at incidents, but does not appear to require a real-time shipment identification system that provides information specific to a particular vehicle. Three years after passage of the Hazardous Materials Transporta- tion Act, DOT was criticized by the National Transportation Safety Board for failing to establish a central data system to provide informa- tion and advice to fire fighters (U.S. Congress. Senate. 1979, 104). A 1979 Congressional Research Service report suggested that DOT could rely on CHEMTREC to provide on-the-scene information to emer- gency responders rather than attempt to create its own system (U.S. Congress. Senate. 1979, 138-139). CHEMTREC is a private, chemi- cal industryâsponsored 24-hr nationwide telephone service that gives emergency responders certain limited information on a hazardous material once the identity of the material is known. CHEMTREC can aid in contacting shippers and manufacturers of materials if the identity of those parties is known. In general CHEMTREC has no ability to obtain rapid access to information about the contents of a specific vehicle.
26 HAZARDOUS MATERIALS SHIPMENT INFORMATION In 1980 DOT entered into a Statement of Formal Recognition with the Chemical Manufacturers Association, the operators of CHEM- TREC. DOT has stated that it believes that the operation of CHEM- TREC meets the intent of the statutory requirement for a central reporting system and data center (Subcommittee on Surface Transpor- tation 1990, 1320-1321). DOT also operates the National Response Center, a 24-hr communications center to which carriers or other responsible parties are required to report certain hazardous materials incidents. However, the purpose of this center is not to provide immedi- ate assistance to emergency responders, but rather to coordinate re- sponse of federal agencies to certain severe releases that require federal involvement in control and cleanup (National Response Center 1992). A proposal for a system designed to provide real-time information about the cargo of a specific vehicle was made during the drafting of legislation regulating hazardous waste in 1984 (Subcommittee on Sur- face Transportation 1990, 1098-1099). Such a provision was not enacted. A 1986 Office of Technology Assessment (OTA) report briefly evaluated the proposal, attributed to a private entrepreneur, John Mulholland of Source Data Network, for "a system to track hazardous waste shipments, which represent less than 1 percent of hazardous materials shipments, in real time." Concerning the extension of this proposal to all hazardous materials shipments, OTA concluded that "online telephone access to real-time information on all hazardous materials shipments is neither feasible nor cost-effective" (OTA 1986, 23). In 1987, Congressman Douglas Applegate of Ohio introduced a bill (H.R. 3334, 100th Congress, Sept. 23) that would have required cre- ation of a central reporting system and data center for hazardous materials shipments. The bill was not enacted. The key provisions were as follows: The system would consist of a central computer data base and communications facilities. "Every generator of hazardous materials and waste" would be required to prepare and transmit to the central facility a bill of lading consistent with the uniform bill of lading already required in federal regulations for hazardous waste shipments. Presumably the shipper, carrier, or receiver would be responsible for seeing that the record of a shipment was entered before the shipment's departure and was up- dated to'reflect transfers between carriers or vehicles and receipt at a
Introduction 27 destination, although the bill did not specify that this updating was to be carried out. The central facility would provide access to this information for police and fire fighters in the event of a wreck, spill, or other incident, along with other information or advice useful in dealing with incidents. Elsewhere, proponents have stated that emergency responders would identify the vehicle they were dealing with to the system by means of the license plate (for a trailer) or the standard identification number painted on the side of all railcars. The legislation itself does not specify how vehicle identification would be accomplished. The system was to be provided by a private entity operating under an agreement with the Secretary of Transportation. It was to be fi- nanced by a $12 fee paid by shippers for each bill of lading entered, with no direct cost to the government for construction or operation of the system. In 1989 and 1990, bills were introduced in both houses to reauthor- ize and amend the Hazardous Materials Transportation Act. These eventually led to passage of the Hazardous Materials Transportation Uniform Safety Act of 1990 (U.S. Congress 1990, 4595-4597), the first substantial revision of the federal regulatory scheme since 1975. During this period, a revised version of the central reporting system and data center proposal was introduced (H.R. 2584, 101st Congress). The proposal dropped the specification of the amount of the fee, recognized the need for some identifying number for the carrier of the shipment that police or fire fighters could determine at an accident scene, and required DOT to establish a program to equip fire fighters and police with telecommunications equipment. During hearings on the hazardous materials reauthorization bills, John Mulholland of Source Data Network asserted that a system for providing on-the-scene identification of hazardous materials ship- ments could be available virtually immediately, using existing private- sector telecommunications and computer capacity. He also stated his belief that such a system could be funded entirely by industry charges, at no cost to the government (Subcommittee on Surface Transportation 1990, 1096-1108). The new legislation was enacted in November 1990. Rather than mandating the central reporting system, Congress called for the Na- tional Academy of Sciences study. Among other provisions relevant to hazardous materials identification, the act requires a comprehensive
28 HAZARDOUS MATERIALS SHIPMENT INFORMATION uniform registration system for hazardous materials shippers and car- riers and a DOT review of ways to improve the existing placarding system. Definition of the Central Reporting System Proposal No definition of the central reporting system has been published or provided to the committee that contains significantly more detailed specifications than those contained in the proposed legislation quoted earlier and in Appendix A. The proposal as it is understood by the committee, together with some assumptions that are essential to fill gaps in the proposal, is as follows. Every shipper of a hazardous material would be required to transmit to a central facility a record of vehicle contents in a prescribed format. The shipper, carrier, or receiver would be responsible for seeing that the record was entered before the shipment's departure and was updated to reflect transfers between vehicles and unloading at a destination. The requirement would apply to shipments for which regulations now require vehicle placarding or hazardous material indication on ship- ping papers, with exceptions for certain materials or quantities. The system would operate nationwide. The central facility would provide access to the record for re- sponders, and possibly other information. Responders would identify a trailer to the system by means of the license plate or by some new required identification number or device. Railcars would be identified by their current identification numbers. The central facility would be financed by a fee paid by shippers or carriers for each bill of lading entered. Since the system would not perform the immediate warning function of placards and possibly could not cover all hazardous shipments, existing requirements for shipping papers and placards would remain in effect. The objective of the central reporting system would be to provide information on vehicle contents in circumstances where shipping pa- pers were destroyed or inaccessible but vehicle identification was visi- ble. The system might also have capabilities for giving information in a more orderly or standardized fashion than the shipping papers, espe- cially in the case of vehicles or trains containing multiple shipments or containers.
Introduction 29 Positions of Interested Parties The proposal for the creation of a central reporting system has at- tracted attention from numerous groups: DOT, other federal agencies involved with hazardous materials as shippers or regulators, local and state fire officials, fire fighters, shippers and producers of hazardous materials, carriers, and potential vendors of devices or services that might be used for shipment information. The positions of some of these groups are summarized in this section. DOT In a May 8, 1990, letter to the chairman of the House Public Works Committee, DOT strongly opposed the central reporting system. The department's objections included the following: The existing information systemâDOT-required shipping papers, package labeling, and placards; the DOT Emergency Response Guide- book (DOT 1990); and CHEMTRECâalready ensures that emer- gency responders at a transportation incident will be able to identify the hazardous materials risks involved. Incidents in which supplying improved or more expeditious haz- ardous materials information to emergency responders would have mitigated the impacts are extremely rare. DOT stated that it had examined the circumstances of all known deaths in 1988 and 1989 resulting from the presence of hazardous materials in transportation incidents and concluded that in no case was failure to identify a cargo a factor in the death. The proposed system could not overcome the problems under the current system caused by shippers or carriers who fail to carry proper placards or documents, because the same shippers and carriers would be unlikely to comply with the electronic filing requirements of the proposed system. The proposed system, which places all reporting requirements on the shipper, would result in a highly incomplete data base because shipments commonly change vehicles between origin and destination. Because of this inaccuracy, together with inevitable misreporting through data entry errors, the data base would constitute a threat to the
30 HAZARDOUS MATERIALS SHIPMENT INFORMATION safety of emergency responders relying on the system for accurate information. The cost of the system would be highâDOT estimated the cost at about $12 billion annually (this estimate was based solely on the $12 per transaction fee proposed in the 1987 bill). The federal government, as a major shipper of hazardous mate- rials, would incur substantial direct costs in complying with the regulation. The bill stipulated that the system be operated by a private entity with little effective oversight. In DOT's interpretation, the bill would require shippers to pay money to a private entity, and the potential for abuse of such an arrangement would be great. Shippers and Producers A group of industry associations representing producers, shippers, and commercial users of hazardous materials, including the Chemical Manufacturers Association, the American Petroleum Institute, the Na- tional Association of Manufacturers, and 19 others, joined in opposing the proposed central reporting system in a 1990 letter to Congress. Their arguments were as follows: In their review of the safety record, they were not able to identify any transportation incident in the preceding S years in which emer- gency responders were killed because the identity of a hazardous mate- rial was not known. The proposed system could create confusion and thus increase risks. The following specific problems were identified: (a) reliance would be placed on an unproven computer system to provide critical information, (b) the proposal ignores the complexities of the trans- actions that occur during transportation of hazardous materials, (c) visible identification numbers on vehicles required by the system would be no more reliable than placards, and (d) the cost would be substantial. Nonetheless, the group stated that the existing hazardous mate- rials identification system can be improved. It suggested (a) establishing a unique vehicle identification numbering system for tank trucks, (b) establishing 24-hr contact points operated by carriers that were capable of identifying the hazardous contents of carriers' vehicles (ship-
Introduction 31 pers are already required to have a 24-hr telephone number on shipping documents), (c) improving existing requirements for placarding and documentation, and (d) improving enforcement of existing regulations. Tracking and identification systems developed by industry to serve its own logistics needs are coming into increasing use. They include satellite tracking of vehicles and shipment-tracking systems that record each stage in the handling of a shipment from origin to destination as it occurs in a computer data base accessible to the shipper or the carrier. Improved information for emergency responders will eventually emerge from these developments, but mandating a specific system now could stall them. The last suggestionâthat the ability for local officials to have access to shipment identification information is likely to evolve as a result of the trend throughout the transportation industry toward development of electronic data interchange and shipment identification systemsâwas made also at the 1989 congressional hearings by a Chemical Manufac- turers Association representative. There is a massive joint industry effort between the Chemical Manufac- turers Association and the [Association of American Railroads] at the present time to deal with the issue of identification of shipments. Stan- dardizing bills of [lading], modifying and developing new and improved methods of rail car identification, such as automatic electronic identifica- tion of rail car numbers, and tieing this in to the railroad systems so that a consist that defines the sequence of cars on a train is accurate with the proper identification of material in the car. And then eventually tie that consist system directly into CHEMTREC so that CHEMTREC, in the case of rail shipments, would have the capability from a rail car number, from a train number, from a location to be able to identify every car on that train, the material in the car, and the location within the train... The CHEMTREC communicators would then have access to that information on his computer terminals within a matter of seconds... The next issue then becomes how do you do the same thing with a bulk truck? And probably more importantly, how would you do this with a barge where you are dealing with large quantities of hazardous material? ... One of the things that is going to have to happen, and we have not begun this initiative yet, but I think will in the future, we have to develop a joint initiative with the American Trucking Associationâor the National
32 HAZARDOUS MATERIALS SHIPMENT INFORMATION Tank Truck Carriers to see what we can do to develop the same communi- cations capabilities for trucks. It is going to be much more difficult but it can be done. (Subcommittee on Surface Transportation 1990, 1062-1064) Carriers Railroad and truck spokesmen have opposed the central reporting system proposal in public statements using the same arguments as the shippers and DOT. Railroads have pointed to their own computer- based car-tracking system, together with their provision of 24-hr access to railroad employees, as fulfilling the need for which the central reporting system is intended. Air freight carriers have special concerns. Identification of hazardous materials on the ground at the scene of an air crash is rarely an issue. However, a major safety concern is the frequent failure of air freight customers to declare and properly label hazardous materials ship- ments. Air carriers fear that the fee for registering with the central reporting system would further discourage proper declaration of haz- ardous air freight. Finally, shippers and carriers can be expected to object to require- ments for the release of business information that could be valuable to competitors into a quasi-public data base. Fire Fighters In congressional testimony, the representative of the International As- sociation of Fire Fighters, a union representing professional fire fight- ers, strongly supported the central reporting system (Subcommittee on Surface Transportation 1990, 896-898, 920-926). The fire fighters believe that they and the public are frequently and unnecessarily ex- posed to heightened risk because of lack of information at incident sites, that the existing system frequently fails to provide adequate information and should be replaced with a computerized shipment identification system, and that the costs of the central reporting system would be reasonable (International Association of Fire Fighters n.d.). Proponents have identified at least five intended benefits of the central reporting system: it would provide emergency responders with cargo information for the vehicles involved in an incident, tie re-
Introduction 33 sponders to a data communications network that would make other data and expert advice and assistance readily available, provide a mechanism to trace hazardous waste shipments to aid enforcement of disposal regulations, provide communities with data identifying the materials that regularly move through their boundaries, and provide a funding source for communications facilities and other emergency response purposes (Muiholland n.d.). Public Emergency Response Officials Opinions of the administrators of fire departments and other state and local emergency response agencies have been split. Some fire chiefs spoke favorably of the central reporting system proposal at hearings (Subcommittee on Surface Transportation 1990,26-33), but the Inter- national Association of Fire Chiefs has not adopted a public position on the issue. Other public safety officials have criticized the proposal as impractical and a diversion from higher priorities (Hermann 1990). STUDY APPROACH To conduct this study, the Transportation Research Board assembled a committee that included experts in hazardous materials safety, freight transportation, information systems, and public policy. The committee did not focus exclusively on a particular system design. Instead, it assessed (a) various potential applications of communications and information technology to aid emergency responders in obtaining in- formation at hazardous materials incidents and (b) nontechnological options for improving information through better regulation, enforce- ment, or training. The committee's approach has been as follows: 1. Identify the information needs of emergency responders: infor- mation requirements to support decisions at incidents; the perfor- mance of existing arrangements for providing that information; the frequency, causes, and consequences of information failures; and the potential benefits of improved information, however obtained. The principal sources for this analysis were case studies of incidents con-
34 HAZARDOUS MATERIALS SHIPMENT INFORMATION ducted for this study and records of incidents compiled by federal and state agencies. Identify technological options that might offer solutions to some of the information problems, including the national central reporting system and systems with other designs. Options were identified from a review of available technologies and examination of systems now in use by shippers and carriers for automated record keeping on shipments and vehicles. The committee considered cost, technical feasibility, and organizational requirements for implementing such systems to the extent that the available data would allow and addressed issues of testing, evaluation, and staging of introduction. Identify nontechnological options for solving information prob- lems, that is, refinements in regulations, improvements in enforcement and training, or provision of new services that do not necessarily depend on applications of computer or telecommunications technol- ogy. Candidate nontechnological improvements were suggested by the incident cases and were also taken from past proposals for reform in regulation, enforcement, and training from emergency response offi- cials, the National Transportation Safety Board, and other sources. It was not possible for the committee to thoroughly evaluate the costs and benefits of each of the options. Rather, the committee identified the major categories of information problems for which the solution may entail regulatory, enforcement, or training measures and identified potential strategies for dealing with each. Finally, the committee decided on recommendations regarding the national central reporting system, a long-term approach to using tech- nology in support of emergency response, and regulatory, enforcement, and training needs. Emergency responder information needs can be divided into two categories: information specific to the vehicles or shipments involved in the incident (e.g., the identity of hazardous materials; quantities, pack- aging, and location in the vehicle; and identities of shipper and carrier) and information that is not vehicle- or shipment-specific (e.g., the hazards of the materials; expert advice on controlling the spill; and population, traffic, or weather at the location of the incident). Al- though this study has considered both kinds of information require- ments, it has concentrated on the need for and benefits of better vehicle- and shipment-specific information and alternative approaches to obtaining vehicle- and shipment-specific information. Provision of
Introduction 35 this information (by requiring real-time reporting of each hazardous materials shipment) is the defining characteristic of the original na- tional central reporting system proposal, and the question of the need for this capability has been the source of most of the controversy over the proposal. Hazardous materials transportation, its regulation, and its safety record are described in Chapter 2; emergency response information needs and the performance of the existing information system are analyzed in Chapter 3; potential technological means of solving infor- mation problems are described in Chapter 4; and nontechnological means are addressed in Chapter 5. REFERENCES ABBREVIATIONS DOT U.S. Department of Transportation OTA Office of Technology Assessment DOT. 1990. 1990 Emergency Response Guidebook. DOT P5800.5. Washington, D.C. Hermann, S. L. 1990. Computerized Hazardous Materials TrackingâAn Idea Whose Time Isn't Now. Environmental Waste Management Magazine. Dec. International Association of Fire Fighters (n.d.). A Nation at Risk: Emergency Response and the Transportation of Hazardous Materials, Washington, D.C. Mulholland, J. F. (n.d.). A Nation at Risk (videotape). National Response Center. 1992. National Response Center. U.S. Coast Guard, DOT, Washington, D.C., March. OTA. 1986. Transportation of Hazardous Materials. Report OTA-SET-304. July. Subcommittee on Surface Transportation. 1990. Hearings. Transportation of Haz- ardous Materials (10142), Hearings Before the Subcommittee on Surface Trans- portation of the Committee on Public Works and Transportation, House of Representatives, USGPO. U.S. Congress. Senate. 1979. Hazardous Materials Transportation: A Review and Analysis of the Department of Transportation's Regulatory Program. April. U.S. Congress. 1990. Hazardous Materials Transportation Uniform Safety Act of 1990, Legislative History.
2 Hazardous Materials Transportation, Regulation, and Emergency Information The general characteristics of haz-ardous materials transportation, its regulation, and the sources of emergency information are reviewed in this chapter. CHARACTERISTICS OF HAZARDOUS MATERIALS TRANSPORTATION Regulated Materials The Hazardous Materials Transportation Act of 1975 (HMTA) and its reauthorizing legislation, the Hazardous Materials Transportation Uniform Safety Act of 1990 (HMTUSA), define a hazardous material as a substance or material in a quantity and form that may pose an unreasonable risk to health, safety, or property when transported in commerce. In setting regulations, the U.S. Department of Transporta- tion (DOT) has classified materials and products on the basis of their hazard characteristics as given in Table 2-1. Thousands of materials and products are covered in the regulations. Many play an important role in the U.S. economy and standard of living. Among the materials designated as hazardous are the common petroleum products diesel fuel, gasoline, and propane; various chemi- kil
Transportation, Regulation, and Emergency Information 37 cals used in agriculture, such as pesticides and fertilizers; explosives and blasting agents used in mining and construction; acids and com- pressed gases used in manufacturing and refining; and numerous con- sumer products, such as paints, alcohols, swimming pooi chemicals, inks, and home cleaning solutions. Also regulated are hazardous wastes and radioactive materials (used by the nuclear energy and medical industries) and infectious substances and disease-causing agents. Shippers, Carriers, and Receivers Because of the large number of regulated materials, shippers, carriers, and receivers of hazardous shipments are diverse and numerous. Major shippers include oil refiners and chemical and bulk gasoline suppliers. Smaller shippers include hospitals, small factories, and residential sup- pliers of home heating fuel. Among gasoline suppliers alone there are some 2,000 large bulk distributors who ship to large manufacturers and utilities and more than 10,000 local distributors who supply individual service stations, farms, and convenience stores (American Petroleum Institute, personal communication). Carriers of hazardous materials also range widely in size, from large interstate rail and trucking companies to smaller short line railroads and local truck operators. Some large shippers operate their own truck fleets and are therefore both shipper and carrier (e.g., petrochemical companies). Whereas some trucking companies specialize in hazardous materials movementâsuch as tank truck operatorsâmost carry both hazardous and nonhazardous cargoes. For example, most less-than- truckload and small-package carriers move hazardous cargoes on a regular basis, although these shipments usually account for only a small share of their business. Receivers of hazardous materials are perhaps most diverse. They include construction sites, retail outlets, farms, hospitals, gasoline sta- tions, printing plants, and waste disposal sites. Water and rail ship- ments are usually terminated (or transferred) at locations where large amounts of hazardous materials are handled, such as ports, junctions, tank farms, warehouses, refineries, factories, and utilities. Receivers of truck shipments are in many of the same locations, but there are many smaller receivers of small shipments or partial loads. For instance, in the United States there are about 150,000 service stations and conve- nience stores that receive gasoline shipments by tank truck (American Petroleum Institute, personal communication).
TABLE 2-1 Hazard Class and Divisions with Example Materials Hazard Definition Class and Name of Class Reference Division , and Division (49 CFR) Examples Brief Description of Hazard 173.50(b)(1) 173.50(b) (2) 173.50(b)(3) 173.50(b)(4) 173.50(b)(5) 173.50(b)(6) 173.115(a) 173.115(a) 1.1 Explosives 1.2 Explosives 1.3 Explosives 1.4 Explosives 1.5 Explosives 1.6 Explosives 2.1 Flammable Gas 2.2 Nonflammable Compressed Gas 2.3 Poison Gas 3 Flammable Liquid Combustible Liquid Black powder Mass explosion Rocket motors Projection hazard Fireworks, Type C Fire with minor blast or projection Squibs Devices with minor explosion hazard Water gels Insensitive article but mass explosion Insensitive article/no mass explosion Propane Compressed oxygen Contents under pressure 173.115(b) Chlorine 173.115(c) Acetone, paint 173.120(b) Diesel Flash point equal to or less than 60.5°C Flash point between 60.5°C and 93°C
4.1 Flammable Solid 173.124(a) Safety matches 4.2 Spontaneously Combustible 173.124(b) Wet cotton 4.3 Dangerous When Wet 173.124(c) Calcium carbide 5.1 Oxidizer 173.127(a) Potassium bromate 5.2 Organic Peroxide 173.128(a) Peroxyacetic acid 6.1 Poisonous Material 173.132(a) Parathion liquid 6.2 Infectious 173.134(a) Virus culture 7 Radioactive 173.403 Uranium-233 8 Corrosive 173.136(a) Caustic soda 9 Miscellaneous 174.140(a) Molten sulfur ORM-D Not Applicable 173.144 Consumer commodity Readily combustible, self-reactive Self-heating materials, ignite or heat when exposed to air Reacts with water to yield flammable or toxic gas or becomes combustible Yield oxygen and fire potential Thermally unstable, burns rapidly, sensitive to impact Toxic to humans Specific activity levels Damages skin on contact or corrodes metal Causes oxygen deprivation Reclassification exception
40 HAZARDOUS MATERIALS SHIPMENT INFORMATION Shipment Sizes and Quantities Hazardous materials are transported in both bulk and nonbulk ship- ments. Bulk shipments, which are defined by DOT as single packagings exceeding 119 gal for liquids, 882 lb for solids, and 1,000 lb for gases, account for most water and rail traffic. Bulk shipments usually consist of liquids or gases transported in tank trucks, tank cars, barge tankers, and intermodal tanks. Tank trucks hold between 2,000 and 10,000 gal, rail tank cars hold as much as 35,000 gal, and barge tankers can hold several hundred thousand gallons. Intermodal tanks, which are trans- ported on flatbed trucks, flat railcars, and barges, can hold approx- imately 5,000 gal. Nonbulk shipments are transported in boxes, drums, cylinders, and smaller packages, often in the same vehicle as nonhazardous cargoes. Estimates by the Office of Technology Assessment (OTA) suggest that about 500,000 shipments of hazardous materials are moved each day in the United States, totaling about 1.5 billion tons (OTA 1986). Shipments included in this figure vary from small packages carried locally to full truckloads and carloads carried across country. Such aggre- gate estimates of hazardous materials shipments, however, provide little insight into potential risk of the traffic, which depends on the types of commodities, packaging, routing, and modes of transport involved. HAZARDOUS MATERIALS REGULATION AND ENFORCEMENT Federal Responsibilities The 1966 Department of Transportation Act consolidated most federal regulatory responsibility for hazardous materials transportation within DOT. Before the act, several federal agencies, including the Interstate Commerce Commission, Department of Treasury, and Civil Aeronautics Board, were responsible for regulating different hazardous materials and their transport modes. HMTA further consolidated haz- ardous materials responsibilities within DOT by creating the Materials Transportation Board (MTB) within the Office of the Secretary. The main objective of the consolidation was to improve regulatory unifor- mity and enforcement by giving central authority to one DOT office to
Transportation, Regulation, and Emergency Information 41 set regulations across modes. The legislation further defined and con- firmed DOT's authority to Determine hazardous materials subject to regulation; Regulate any traffic "affecting" interstate commerce, thereby cov- ering some intrastate traffic; Regulate shipper handling, labeling, and packaging of hazardous materials; Regulate manufacturers, reconditioners, and testers of shipping containers for hazardous materials; Conduct inspections and penalize violators; and Preempt state, local, and Indian tribal rules that were found to be inconsistent with federal regulations. MTB was later moved to the newly created Research and Special Programs Administration (RSPA) and renamed the Office of Haz- ardous Materials Safety, which is currently responsible for issuing most hazardous materials regulations. However, the DOT modal adminis- trationsâincluding the Federal Railroad Administration (FRA), Federal Aviation Administration, U.S. Coast Guard, and Federal High- way Administration (FHWA)âhave retained important roles, pri- marily in regulatory implementation and enforcement. FI-IWA, for instance, inspects motor carrier and shipper facilities and provides funds to states to enforce motor carrier regulations. Likewise, FRA has responsibility for ensuring that rail carriers and shippers understand and comply with the regulations. Several federal agencies in addition to DOT have a role in regulating and enforcing various aspects of hazardous materials transportation. For instance, the Environmental Protection Agency (EPA) designates hazardous substances and wastes and requires manifesting of waste shipments from their origin to their disposal. In doing so, EPA has recognized DOT requirements but has set more stringent notification, marking, and shipping paper (manifest) requirements. Other federal agencies with responsibilities for aspects of hazardous materials trans- port include the Nuclear Regulatory Commission, which sets standards for the design and performance of packages used to move certain radioactive materials, and the Department of Defense, which has adopted restrictions and procedures for hazardous materials move- ments by the military (although most hazardous materials shipments by private defense contractors are subject to DOT regulations).
42 HAZARDOUS MATERIALS SHIPMENT INFORMATION Within the Department of Labor, the Occupational Safety and Health Administration (OSHA) is responsible for regulating hazardous materials used and stored in the workplace. Accordingly, the safety of transportation workers handling hazardous materials is within OSHA's responsibility. Some of these responsibilities are handled by DOT according to a memorandum of understanding between the two federal agencies. In addition, OSHA has package marking and labeling requirements for materials it has designated as hazardous in the work- place, and it requires manufacturers and shippers of OSHA-designated hazardous materials to provide employers with Material Safety Data Sheets (MSDSs) that contain emergency response information. State and Local Roles A primary responsibility of state and local governments is enforcement of hazardous materials regulations, especially those pertaining to truck transport. The federally funded Motor Carriers Safety Assistance Pro- gram (MCSAP), which is administered by FI-IWA, has increased unifor- mity of state regulation of motor carriers and enforcement by providing funds for enforcement. MCSAP encourages states to conduct more frequent roadside and terminal inspections to ensure that DOT regula- tions are followed, including those pertaining to hazardous materials. Historically, states have had less authority to enforce hazardous mate- rials regulations for railroads, but recent changes in FRA provisions have given states broader authority to enforce these federal regulations as well. In the regulation of hazardous materials transportation, federal reg- ulations preempt most state and local authority. In particular, state and, local governments cannot adopt requirements that unreasonably bur- den interstate commerce, reduce the overall safety of the transportation system, or interfere with the uniformity of federal regulatory standards (for instance, by developing different placard symbols). HMTA gave DOT explicit authority to regulate any traffic "affecting" interstate commerce. Hence, whereas state and local governments can designate hazardous materials transportation routes for clear safety reasons (e.g., to restrict movements on certain bridges and in certain tunnels), they have more limited authority to impose permits and fees for hazardous materials transport or to adopt hazardous materials regulations that differ from those of the federal government.
Transportation, Regulation, and Emergency Information 43 Historically, states have had the greatest freedom to regulate intra- state highway transportation, although DOT has encouraged states to adopt the federal regulations. In recent years, most states have adopted the federal regulations in whole or large part. The movement toward greater federal and state uniformity in hazardous materials regulation was accelerated by HMTUSA, which requires DOT to extend federal hazardous materials regulations to intrastate motor carriers. Major HMTUSA Provisions HMTUSA has several provisions that amend HMTA. Among the most relevant to this study are the following: DOT must extend hazardous materials transport regulations to intrastate (including highway) and international transport. DOT must set standards for states to use in designating hazardous materials routes for highway transport. Motor carriers must obtain safety permits if they carry certain radioactive materials, explosives, toxic inhalants, or liquid natural gases. Shippers and carriers of bulk and nonbulk placardable shipments of hazardous materialsâas well as shippers and carriers of certain explosives, radioactive materials, and material toxic by inhalationâ must register with DOT. DOT must determine methods to improve placarding, establish a central reporting system for hazardous materials incidents and a com- puterized telecommunications center, and assess the feasibility of requiring carriers of hazardous materials to provide continuously mon- itored emergency telephone systems. Special funding is provided for DOT to increase the number of federal hazardous materials inspectors by 30, allocated among FRA, FHWA, and RSPA. DOT must establish a planning and training grant program for emergency response, including curriculum development. DOT must also establish training requirements to be provided by hazardous mate- rials employers to their employees, including truck drivers and train crews. Minimum civil penalties of $250 are established for hazardous materials violations and maximum penalties are increased from
44 HAZARDOUS MATERIALS SHIPMENT INFORMATION $10,000 to $25,000. Maximum criminal penalties ($500,000 or 5 years in prison, or both) are also established. To comply with these requirements, DOT is establishing new regula- tions and programs. Examples of new rule making (at various stages of completion) covering some of these HMTUSA provisions include docket numbers HM-126F (hazardous materials employee training), HM-206 (central reporting system, 24-hr carrier telephone num- ber, and improved placarding), and HM-208 (shipper and carrier registration). EMERGENCY RESPONSE INFORMATION Federal regulations require shippers and carriers of hazardous mate- rials to communicate the hazards of their shipments by providing placards, shipping papers, and package markings and labels (49 CFR 100-199). These requirements, which are part of the current system of hazard identification (see accompanying text box), are intended to provide essential information about hazardous cargo to the public and emergency response personnel when incidents occur. Other informa- tion sources include the DOT Emergency Response Guidebook, MSDSs, and CHEMTREC. Placards Placards are diamond-shaped signs placed on the ends and sides of trucks, railcars, and intermodal containers. Four placards per vehicle are required (front, rear, and two sides). Through coded colors and symbols, placards warn of the presence of hazardous cargo and identify the class of hazard present (e.g., flammable, corrosive, or poison) (Figure 2-1). When hazardous materials are transported in bulk ship- ments (e.g., by tank cars and tank trucks), the four-digit United Na- tionslNorth American (UN/NA) material identification code must be displayed on the placard or accompanying orange panels. The UN/NA number can be cross-referenced with response information contained in the DOT Emergency Response Guidebook. DOT has developed tables indicating which placards correspond to individual hazard classes and divisions (Table 2-2). "Table 1" mate-
Current DOT Hazard Identification and Emergency Response Information System Shipping Paper Requirements Number and types of packages Gross weight of each hazardous material Proper shipping name, hazard class, UN/NA number, and packing group Technical names for N.O.S. and other specified generic materials Additional entries for poisons, reactive materials, dangerous-when-wet materials, cargo aircraft only shipments, exemptions, limited quan- tities, empty packagings Reportable quantity entry for hazardous substances Immediate emergency response information (may be attached to ship- ping paper or referenced to Emergency Response Guidebook on board the vehicle) consisting of basic description and technical name of the material; immediate hazards to health; risks of fire or explo- sion; precautions to be taken in the event of an accident or incident; methods for handling fires, spills, or leaks in the absence of fire; and preliminary first aid measures 24-hr emergency response telephone number of person having compre- hensive knowledge of materials in the shipment Placard information for rail shipments Shipper's certification Package/Container Marking Requirements, Less Than Bulk Proper shipping name Technical names for N.O.S. and other specified generic materials ID number (UN/NA number) Reportable quantity if appropriate Name and address of consignor or consignee Package certification/specification Exemption packaging number if applicable Additional markings for elevated temperature materials, LX number for certain explosives, ORM-D designation, orientation arrows for liq- uids in combination packagings Package/Container Marking Requirements, Bulk Material ID number on placards or orange panels Proper shipping number for certain materials Additional markings as appropriate Package certification/specification (continued on next page)
46 HAZARDOUS MATERIALS SHIPMENT INFORMATION Current DOT Hazard Identification and Emergency Response Information System (continued) Labeling Requirements If limited quantity exceeded per package Specific labels per hazard class Size, color, symbols, and so forth are specified Applied to surface of package Multiple labels required for some materials Additional labeling requirements if subsidiary hazard, cargo aircraft only, empty, spontaneously combustible, dangerous when wet, or keep away from food Placarding Requirements Four placards (front, rear, and two sides) per vehicle For any quantity of certain explosives, poison gases, poisons, radioac- tives, and dangerous-when-wet materials For all materials if the aggregate gross weight exceeds 1,000 lb For each material if weight exceeds 5,000 lb Multiple placards if appropriate Orange panels or placards with ID numbers for bulk shipments Some residues must be placarded Size, color, symbols, and so forth are specified Emergency Response Guidebook Provided to all emergency responders Guidebook (or equivalent) required to be on board each vehicle and at each shipping, receiving, storage, and other location where haz- ardous materials are handled incident to transportation rialsâwhich consist of certain explosives, poison gases, radioactive materials, and materials that are dangerous when wetâare considered the most hazardous and must always be placarded in transportation. "Table 2" materials, which are considered less hazardous, do not require placards if transported in quantities of 1,000 lb or less. Also, when two or more shipments of these materials are transported in a vehicle and none of the individual shipments (loaded at one location) weighs more than 5,000 lb (and the aggregate weight of all the haz- ardous materials in the vehicle exceeds 1,000 lb), a Dangerous placard may be substituted for the specific placard for each hazard class.
NGEROO FIGURE 2-1 Examples of placards. Top, Combustible placard (red with white symbol); middle, Corrosive placard (center and lower areas black except for letters, which must be white); bottom, Dangerous placard (red upper and lower triangle).
TABLE 2-2 Placard References (49 CFR 172) Placard Design Hazard Class Section and Division Placard Notation Reference (49 CFR) Table 1 1.1 Explosives 1.1 172.522 1.2 Explosives 1.2 172.522 1.3 Explosives 1.3 172.522 2.3 Poison gas 172.540 4.3 Dangerous when wet 172.548 6.1 (PG 1, inhalation Poison 172.554 hazard only) 7 (Radioactive Yellow Radioactive 172.556 III label only) Table 2 1.4 1.5 1.6 2.1 2.2 3 Combustible liquid 4.1 4.2 5.1 5.2 6.1 (PG I or II, other than PG I inha- lation hazard) 6.1 (PG III) 6.2 8 9 ORM-D Explosives 1.4 Explosives 1.5 Explosives 1.6 Flammable gas Nonflammable gas Flammable Combustible Flammable solid Spontaneously combustible Oxidizer Organic peroxide Poison Keep away from food None Corrosive Class 9 None 172.523 172.524 172.525 172.532 172.528 172.542 172.544 172.546 172.547 172.550 172.552 172.554 172.554 172.558 172.560
Transportation, Regulation, and Emergency Information 49 Shipping Papers Most shipments of hazardous materials must be accompanied by shipping papers that describe the hazardous material and contain certification by the shipper that the shipment conforms with all DOT hazardous materials regulations. Instructions for describing haz- ardous cargoes in shipping paper entries are provided in DOT regu- lations. Entries must include the material's quantity (gross weight and number and type of packages), proper shipping name, hazard class or division, UN/NA number, and packing group, as identified in the DOT Hazardous Materials Table. If a specific shipping name is not listed in the DOT table, a shipping name must be selected from general description and n.o.s. (not otherwise specified) entries that correspond to the hazard class of the material. Under these circumstances, techni- cal names of the components causing the hazard must be entered in parentheses. Shipping papers must also list the name and address of the shipper or receiver and an emergency telephone number that can be used by responders to obtain more detailed emergency information about the shipment. Basic hazard and response information must be entered on the shipping papers or attached to them. The information must include immediate hazards to health, methods of handling fires and spills, risk of fire and explosion, precautions in the event of an incident, and preliminary first aid measures. The DOT Emergency Response Guide- book, which must be in the vehicle, is often used for this purpose. The MSDS may also be used if the document contains the information required by DOT. Shipping papers must be placed in the transport vehicle in an access- ible location. In cases involving mixed loads of both hazardous and nonhazardous materials, shipping papers for hazardous materials must appear first or be distinctively tabbed so that they can be accessed quickly by the driver and emergency responders. In most cases, DOT does not specify the use of a standard document, and therefore required entries can be provided on shipping documents commonly carried in the vehicle for business purposes. The exception is hazardous waste shipments, which must be accompanied by a special hazardous waste manifest that lists special shipper, carrier, and receiver identification numbers required by EPA. Trucking companies refer to shipping papers as "bills of lading" (generated by the shipper) or "freight bills" (gener- ated by the carrier). Railroads use "waybills" to list emergency infor- mation on hazardous cargoes in individual cars (each car has its own
SO HAZARDOUS MATERIALS SHIPMENT INFORMATION waybill). Most major railroads have standardized their train consist to include hazardous materials information. The position of each car, including those containing hazardous materials, is listed in the train consist. However, if cars are added or dropped at intermediate points en route, the consist may not accurately list the position of hazardous materials cars in the train. Labels and Markings on Containers and Packages Like placards, most labels contain symbolic and numeric representa- tions of the hazards associated with a particular class or division of hazardous materials. Labels are required on most hazardous materials packages and containers, except for shipments of very small quantities. In some cases, special handling labels must also be attached to the packaging (for instance, if the material is not safe for transport on passenger aircraft or must be separated from food shipments). Some shipments may require multiple labels if the shipment contains more than one hazard class or division. Additional markings must accompany labels on nonbulk and some bulk shipments. They include the proper shipping name of the material, its UN/NA number (on the placard or orange panel for bulk ship- ments), name and address of the shipper or receiver, package certifica- tion information, and certain additional information as required (e.g., package orientation markings and EPA-reportable quantities). Recent Changes in Information Requirements For the past several years, DOT has been revising federal hazardous materials regulations to simplify them, reduce their volume, and make them more consistent with international standards. Major revisions are contained in RSPA Docket HM-181, issued as a final rule in December 1991. The HM-181 revisions, which are now being implemented, are the most comprehensive regulatory changes to date. Because of changes in hazard classifications, emergency information requirements are affected for some domestic and international shipments. For in- stance, HM-181 creates the new labels and placards Spontaneously Combustible, Keep Away from Food, and Class 9. It also revises the scheme for classifying hazards by creating several new hazard divisions within hazard classes. The new classifications will require changes in
Transportation, Regulation, and Emergency Information 51 shipping descriptions for all hazardous materials as well as marking, labeling, and placarding changes for some materials. Because of the large number of regulatory changes, a transitional period has been established by DOT to allow industry time to comply. Some changes became effective in October 1991, but most were given transitional compliance periods ranging from 1 to S years after the effective date. Attempts were made in this study to ensure that all references to DOT hazardous materials regulations reflect the new regulatory changes. Other Information Sources Most state and local police and fire departments have copies of the DOT Emergency Response Guidebook. The guidebook provides basic hazard and response information for those who are first to arrive at the scene of an incident. Responders can match placards on vehicles with those illustrated in the guidebook for preliminary response and first aid advice. The information in the guidebook is meant to be augmented by expert technical advice. One mechanism for obtaining this advice is CHEMTREC, a 24-hr telephone hotline provided by the Chemical Manufacturers Association. Using a toll-free number, responders can call CHEMTREC to obtain hazard information and technical guidance by giving the UN/NA number or the name of the product and nature of the problem if the number is not found. CHEMTREC can also estab- lish a communications link between responders on the scene and the shippers (if known) of the materials involved. CHEMTREC maintains an on-line data base on the chemical, physi- cal, and toxic properties of hundreds of thousands of products of participating chemical companies. Much of this information is ob- tained from MSDSs submitted by shippers to CHEMTREC. The MSDS is a widely used reference document because it contains detailed information on the material's physical properties, major health haz- ards, material incompatibilities, and emergency handling methods. Training and Preparation To Use Information Sources First responders to an incident are often local police and fire units, whose only warning of hazardous materials may be the presence of
52 HAZARDOUS MATERIALS SHIPMENT INFORMATION placards. Most emergency responders are trained to recognize placards and take initial protective measures, but few are trained at a higher level of competence. The strongest preparation is in large communities, which often have special hazardous materials teams and units. Prepara- tion for hazardous materials incidents is usually weakest in rural areas, where local fire departments are often manned by volunteers who may not have hazardous materials training or equipment. The International Association of Fire Chiefs (IAFC) has estimated that the total fire service population in the United States exceeds 1 million, but that about 85 percent are part-time employees or unpaid volunteers (OTA 1986; IAFC 1985). States play a critical role when incidents occur in rural areas. Some states have established hazardous materials response teams to assist in major emergencies. In addition, state police, fire marshals, emergency management agencies, and environmental and health agencies respond to major incidents. If the incident occurs at or near rail yards or terminals, railroad personnel may assist with the response. Because they are large and operate over fixed routes, railroads are more likely than trucking companies to have personnel and equipment available for this purpose, and some have prepared to coordinate response activities with local communities. Many large shippers can also provide emergency response teams. For instance, the chemical industry maintains a national chain of emer- gency response teams (CHEMNET) that can be deployed in a few hours to chemical emergencies around the country. In addition, since passage of the federal Superfund Amendment and Reauthorization Act in 1986, many local governments and industries have developed Local Emergency Planning Committees and mutual-aid networks that can provide special equipment and response personnel to hazardous mate- rials transportation incidents. At the federal level, the U.S. Coast Guard operates the National Response Center, which acts as the central coor- dinator of federal agencies that need to be notified of incidents. Also, both the Coast Guard and EPA can provide on-scene coordinators to monitor incidents and provide technical advice and guidance if necessary. Several states (e.g., Ohio, Virginia, and Tennessee) have established fire institutes and full-time hazardous materials response teams to train local officials and assist in major hazardous materials transportation emergencies (TRB 1988, 80-81). As mentioned previously, HMTUSA provides funding for state and local training of emergency responders
Transportation, Regulation, and Emergency Information 53 and for development of course curriculum. DOT will provide grants to states that apply for federal training funds. In addition, HMTUSA requires DOT to ensure that transportation workers who handle haz- ardous materials, including train crews and truck drivers, are trained for safety and to understand their role in the event of an incident. To comply with this requirement, DOT has issued regulations requiring hazardous materials shippers and carriers to train and test employees in hazardous materials awareness and safety (Docket HM-126F). The value of federal assistance for training of local response person- nel depends on existing state and local response capabilities. For several years, RSPA has offered hazardous materials response and safety train- ing through its Transportation Safety Institute to federal, state, and local emergency response personnel and to transport workers who handle hazardous materials. RSPA has also developed a number of training modules consisting of workbooks and audiovisual presenta- tions that demonstrate how hazardous materials information sources should be used in an emergency. The training modules are available to emergency responders and transportation workers. Traditionally, the primary federal providers of emergency response training have been the Federal Emergency Management Administra- tion's National Fire Academy and Emergency Management Institute and EPA's Environmental Response Program. These programs, which cover hazardous materials incident mitigation, command, and contain- ment procedures, provide training to state and local officials through resident, field, and teleconference courses that are funded by a combi- nation of federal, state, and local sources. REFERENCES ABBREVIATIONS IAFC International Association of Fire Chiefs OTA Office of Technology Assessment TRB Transportation Research Board IAFC. 1985. Hazardous Materials Teams Survey: Summary of Results. Washing- ton, D.C. OTA. 1986. Transportation of Hazardous Materials. Report OTA-SET-304. Wash- ington, D.C. TRB. 1988. Special Report 219: Pipelines and Public Safety. National Research Council, Washington, D.C.
ki Necessity and Benefits of Improved Information Before considering options for im-proving emergency information, it is necessary to examine the performance of the existing information system, both to recognize the kinds of information problems that occur and to identify problems that are likely to yield significant benefits if resolved. As a means of doing so, Chapter 3 is organized around the following four questions: 'What are the most critical information needs of fire fighters, police, and other emergency responders at hazardous materials incidents? What are the most common problems encountered by responders when depending on existing information sourcesâsuch as placards, shipping papers, and package labelsâto meet these needs? How often do information problems occur? What are the costs associated with these problems and, therefore, the potential benefits of resolving them? After a brief explanation of the data used in the chapter, each of the four questions is addressed. First, the most critical information needs of responders are considered. Because the circumstances of each haz- ardous materials incident will differâas do responder resources, prac- tices, experience, and trainingâthe information needed will vary by incident and community. Nevertheless, certain kinds of information, 54
Necessity and Benefits of Improved Information 55 such as initial warnings about the presence of hazardous materials and the identity of the materials involved, are required and expected by responders at virtually all kinds of hazardous materials incidents. This information helps in preventing injuries to first responders and in obtaining more detailed information, such as the physical properties of the materials and guidance on protective equipment and first aid. A second section discusses how existing information sources do not always meet the critical information needs of responders. Actual cases are examined to illustrate the kinds of problems that occur. Evidence from the case studies also provides a rough indication of the share of incidents with information problems. In a third section, the evidence is combined with estimates of total hazardous materials incidents to predict the frequency of information problems on a national basis. Available data are not well suited for this purpose, making conclusive estimates impossible. The costs, or adverse consequences, of information problems are con- sidered in a fourth section. The discussion emphasizes that fatalities and injuries are not the only potential costs of poor information and that information failures can contribute to other consequences, such as expo- sure of response personnel, unnecessary evacuations, prolonged traffic delays, and inefficient use of response resources. Cost estimates are devel- oped for some of these consequences to indicate their potential magnitude and illustrate the possible benefits of information improvements. The focus of the chapter is on hazardous cargoes moved by rail and truck, which account for the majority of incidents, followed by air and water. Because hazardous materials are moved by rail and truck throughout the country, the safety of these two modes is of concern to the largest number of emergency responders and is therefore central to the consideration of a national shipment information system or other options for improving emergency information. Although the analyses and discussion in the chapter do not address water and air transport directly, many of the conclusions are relevant to these modes. DATA SOURCES To help answer the questions posed, two types of data are required. The first is descriptive statisticson hazardous materials incidents. Statistical data are necessary to determine the frequency of incidents, their gen- eral characteristics (such as the types of shipments involved), and their consequences (such as fatalities, injuries, and evacuations). The second
56 HAZARDOUS MATERIALS SHIPMENT INFORMATION is evidence on, the performance of the existing information system in meeting the needs of responders during incidents. The availability and reliability of both types of data are limited. No comprehensive national census or representative sample of hazardous materials incidents exists. Similarly, there has never been a systematic study of the information needs of responders or the adequacy of the existing system in meeting these needs. It was necessary for the committee to gather its own, sometimes sketchy, evidence from a number of sources. Hence, references are made throughout the chapter to incident statistics and case studies developed for this study. Each of the data sources is described in the following subsections. Further discussion of the data used by the com- mittee, the kinds of data needed, and suggestions for improving the data is presented in Appendix B. Incident Statistical Data For the purposes of this study, the hazardous materials incidents of most concern are those involving public responders, who are the pri- mary users of the emergency information system and the direct bene- ficiaries of system improvements. Public responders are called to a wide variety of incidents, ranging from major tank car derailments to minor or suspected spills from small shipments at terminals and loading facilities. No single data source covers all such incidents. 'Whereas many data bases are suitable for the purposes for which they were developed, they are seldom well suited for comparison and aggregation with other incident data bases because of differences in reporting requirements, incident definitions, and data completeness. After re- viewing the available data and recognizing their limitations, the com- mittee settled on the following three data sources to provide some rough statistics on hazardous materials incidents: Research and Special Programs Administration (RSPA) incident data: The only national data base aimed specifically at hazardous materials transportation incidents is RSPA's Hazardous Materials In- formation System (HMIS). Since 1971, RSPA has required interstate carriers to file reports of incidents that involve an unintentional release of hazardous materials during transportation, including loading, un- loading, and temporary storage en route.1 Figure 3-1 shows the number
Necessity and Benefits of Improved Information 57 Reported Incidents (Thousands) o?= 82 83 84 85 86 87 88 89 90 91 Year Air/Water Truck RaiI FIGURE 3-1 RSPA incident data by year and mode, 1982 to 1991 (source: RSPA). The data are not indicative of the volume of hazardous materials incidents handled by public-sector emergency responders, which are most relevant to this study. Estimates of truck and rail incidents involving public-sector, emergency responders are presented in Table 34. of incidents reported to RSPA during the past 10 years by transport mode. The RSPA data are referenced in this study primarily to develop estimates of rail incidents. Most rail shipments are carried by major interstate carriers, which are required to submit incident reports to RSPA. HMIS has several significant shortcomings that preclude its use in estimating truck incidents. In particular, intrastate carriers and local truck operators are not required to submit incident reports, potentially resulting in thousands of unreported incidents. Also, HMIS is domi- nated by reports of minor spills incurred by a small number of large interstate trucking companies at their terminals. For example, a corn-
58 HAZARDOUS MATERIALS SHIPMENT INFORMATION mittee review of 1990 RSPA data indicates that one-third of truck incidents were reported by three of the country's largest less-than- truckload (LTL) and small-package trucking companies. Most of these incidents are of secondary concern to this study because they do not involve public responders.' California state incident data: In recent years, growing awareness of hazardous materials transportation by both the public and emer- gency responders has caused several states to start collecting incident data. Most state data bases were started within the past S years. All rely on reports from public agencies rather than carriers, and most include both fixed-site and transportation incidents. Because the incident re- ports are submitted by public responders, they are especially relevant to this study. However, differences in reporting criteria and compliance among states make it difficult to compare and aggregate incident statistics across states.' The committee determined that for the pur- poses of this study the most suitable state data base is the California Hazardous Materials Incident Reporting System (CHMIRS). CHMIRS reports are filed by an estimated two-thirds of all county and municipal fire and police departments in the state (California Office of Emergency Services, personal communication). Given the shortcom- ings of the RSPA data for truck incidents as well as the focus of this study on incidents involving a public response, the CHMIRS data were used in this study as the primary source of truck incident data. CHEMTREC notification data: CHEMTREC maintains statisti- cal records on incident notifications received from carriers, shippers, and government agencies, including state and local fire and police departments. Because CHEMTREC notification is not required, the records are incomplete. Nevertheless, most railroads routinely notify CHEMTREC of incidents because shippers, who are their customers, often request that they do so (Chemical Manufacturers Association, personal communication). Accordingly, the CHEMTREC data are most reliable for rail incidents and are used in this chapter along with the RSPA data to develop rail incident estimates. A more complete discussion of available data sources is provided in Transportation of Hazardous Materials (OTA 1986). Some of these sources were used to verify estimates derived from the RSPA, CHMIRS, and CHEMTREC data. For the past 17 years, the National Fire Information Council, under the auspices of the United States Fire Administration of the Federal
Necessity and Benefits of Improved Information 59 Emergency Management Agency, has implemented a voluntary inci- dent reporting system for state and local fire agencies, known as the National Fire Information Reporting System (NFIRS). NFIRS was developed as a result of the Federal Fire Prevention and Control Act of 1974 to provide a comprehensive picture of fire incidents across the United States. NFIRS reports are submitted to the U.S. Fire Adminis- tration through state agencies using computer software. Recently, a more detailed hazardous materials reporting component has been ad- ded to the system. Since its first full year of implementation in 1991, 14 states have started using the hazardous materials component (most have submitted partial data only). As more states participate, NFIRS should become another valuable source of data on hazardous materials transportation incidents. Performance Data Data on the performance of the information system are even more limited than incident data. For instance, HMIS does not indicate whether local police or fire departments responded to the incident, nor does it indicate whether information sources required by the U.S. Department of Transportation (DOT), such as placards and shipping papers, were present and effective in convcying hazard information.4 Likewise, none of the state incident data provides this kind of informa- tion in sufficient detail. In reviewing available data, the committee found National Transportation Safety Board (NTSB) investigations of truck and rail incidents to be the most useful source of performance data. As a standard practice, NTSB examines the adequacy of emer- gency response operations during its investigations of major hazardous materials incidents. On the basis of these investigations, NTSB has made dozens of recommendations during the past 10 years concerning needed improvements in emergency information (Appendix Q. How- ever, because NTSB investigations are few in number and seldom focus on emergency response operations, they do not provide sufficient evi- dence to evaluate the performance of the existing information system. Special Case Studies The limited availability of data compelled the committee to conduct incident case studies to evaluate the existing emergency information
60 HAZARDOUS MATERIALS SHIPMENT INFORMATION system. The case studies are aimed at answering the questions posed at the beginning of the chapter. Their primary purpose is to identify the critical information requirements of emergency responders and illus- trate the kinds of problems encountered in the field. Their secondary purpose is to estimate the frequency with which information problems occur on a national basis and identify significant costs associated with the problems. The case studies consist of the committee's firsthand examination of 40 hazardous materials truck incidents in California and its sec- ondhand examination of 85 previous NTSB incident investigations. The cases were selected for study without prior knowledge of whether information problems were encountered by responders during the inci- dent. A total of 125 cases were examined, including 80 truck and 45 rail incidents. The cases were derived as follows: California truck cases: The committee examined 40 hazardous materials truck incidents that occurred in California during 1990 and 1991. The incidents were reported initially in the RSPA and CHMIRS data bases. California was selected for the case studies because it is a large and diverse state with considerable hazardous materials traffic and because the combined CHMIRS and RSPA data bases provided a relatively large pool from which to select cases. For each truck incident, copies of the original RSPA or CHMIRS reports were obtained and persons involved in the response were contacted by telephone and questioned about the availability of information during the incident. Newspaper accounts of the incident were also reviewed when available. NTSB truck cases: The California cases were supplemented with 40 additional cases derived from an NTSB study of 189 large-truck crashes (NTSB 1988). The cases consist of all incidents in the NTSB study that involved trucks carrying hazardous cargoes. The original purpose of the NTSB study was to determine the cause of each crash in order to identify recurring causes. NTSB routinely evaluates the perfor- mance of emergency response operations during its investigations. NTSB investigators personally visited the incident sites and interviewed the emergency responders involved. For each case, the committee was able to review the results of the NTSB investigation of emergency response operations from backup documentation on microfiche. The documentation was used to identify cases in which emergency re- sponders encountered information problems and to determine major consequences of these problems. This documentation was supple- mented by a review of newspaper accounts of the incidents.
Necessity and Benefits of Improved Information 61 NTSB rail cases: In 1991, NTSB completed a major study of 45 hazardous materials rail incidents that occurred between March 1988 and February 1989 (NTSB 1991). NTSB did not investigate the cases to determine the adequacy of emergency response information, but in each case NTSB investigators documented the chronology of the re- sponse, including the availability and effectiveness of placards, ship- ping papers, and other sources of emergency information. As with the NTSB truck cases, the backup documentation for each rail case was obtained by the committee and reviewed, along with newspaper ac- counts and other written reports of the incidents. The documentation was used to identify cases in which information problems occurred and major consequences of these problems. All 125 cases had "consequential" outcomes, which are defined for the purposes of this study as involving a fatality, reportable injury, evacuation, or major highway or track closing. Combined, the cases involved 260 reported injuries (24 fatal and 236 nonfatal), 49 work- place and public evacuations, 27 fires and explosions, 23 major highway closings, and 34 derailments.5 Consequential incidents were selected for examination because they tend to be better docu- mented than other incidents and are of most,concern to responders and the public. However, severe consequences are not typical of most hazardous materials incidents, which seldom involve major wrecks, fires, injuries, public evacuations, or other serious outcomes. For exam- ple, about 1 in 25, or 4 percent, of the incidents reported to CHMIRS (from which California truck cases were selected) had one or more of these consequences. Many of the adverse consequences in the case studies were the direct result of the crash, derailment, or other event that led to the incident and were unrelated to the hazards of the materials being transported. A summary of the 125 truck and rail cases is provided in Appendix B. Limitations of the Case Studies Although the case studies provide some helpful insights, they have limitations that make their results tentative. The following are the most critical: Eighty-five ,f the case studies were developed almost exclusively from secondhand analysis of documentation from previous NTSB
62 HAZARDOUS MATERIALS SHIPMENT INFORMATION investigations. Because none of the NTSB investigations focused ex- clusively on emergency information, some subtleâyet importantâ information problems may not have been recorded in the incident documentation. Most of the California truck cases were limited to a review of brief incident reports (submitted to RSPA and CHMIRS) and follow-up telephone interviews with one or two of the responders involved. Personal interviews with responders shortly after the incident would have increased the likelihood that events were recalled accurately. Because of the potential for responders to forget incident details over time, some information problems may have been overlooked during the case studies. The reported consequences of the 125 incidents examinedâsuch as injuries, property losses, and evacuationsâmay not have been com- plete or precise. Incident consequences were determined primarily from RSPA and CHMIRS reports and NTSB documents, which were supplemented with details from responder interviews and available newspaper reports. As a result, some consequences, such as delayed injuries and long-term illnesses, may not have been reported and are therefore not included in the case study examinations. The NTSB-derived cases tend to have more serious consequences in terms of fatalities and injuries than the California cases. NTSB usually investigates major incidents; hence, these cases often involve severe crashes and derailments. The California truck cases consist of a wider variety of incidents, including many that did not involve crashes or other exceptional circumstances. Consequently, the California cases may be more representative of routine incidents, whereas the NTSB- derived cases may represent the most serious kinds of cases.' The need for better data is stressed throughout the chapter. Nev- ertheless, the committee was compelled to rely on the findings from the case studies. Better data could not be collected or analyzed within the scope and time frame of the study. Despite their limitations, the com- mittee found the case studies essential to understanding many issues. Other Supplemental Incidents Because the 125 case studies cannot reveal the variety of information problems encountered by emergency responders at hazardous mate-
Necessity and Benefits of Improved Information 63 rials incidents, 12 other major incidents (8 truck and 4 rail) that illustrate information problems are cited in the chapter and summa- rized in Appendix B. The incidents were derived from newspaper reports and NTSB investigations. Many were highly publicized at the time. Because they were selected subjectively, they are cited for illustra- tive purposes only and are not used to estimate the frequency of information problems. Responder Interviews Finally, as an additional information source, interviews were con- ducted with a number of emergency responders, including state and local fire fighters and industry hazardous materials handling and re- sponse experts.7 The responders were asked to identify the kind of information they require at hazardous materials incidents and the types of information problems they encounter. The interviews were partic- ularly helpful in determining the information needs of responders. CRITICAL INFORMATION NEEDS OF RESPONDERS What are the most critical information needs of fire fighters, police, and other emergency responders at hazardous materials incidents? During the interviews, responders were asked to identify the types of information required at hazardous materials incidents. Although a wide range of needs were identified, the three discussed in this section were consistently characterized as most significant. Each is vehicle- or shipment-specificâthat is, obtained primarily from information sources accompanying the vehicle or shipmentâand must be provided before more detailed information can be sought to safely and efficiently handle the incident. 1. Forewarning of the presence of hazardous materials: Fire fighters, police, and other emergency responders may not be aware of the presence of hazardous materials when arriving at an incident. Early and clear warning is the first and most critical type of information required, because it can prevent first responders from approaching the site unprepared (e.g., without personal protection).
64 HAZARDOUS MATERIALS SHIPMENT INFORMATION Initial warning information is the primary function of placards (Ta- ble 3-1). The prominent display of the diamond-shaped placard is intended to immediately warn responders and bystanders that hazards are present and thereby reduce the chance of someone inadvertently entering the incident site. This function is served whether or not accom- panying symbols, coded colors, and numbers (e.g., hazard class and material identification numbers) are recognizable. The presence of placards, if recognized, also allows first responders to alert dispatchers of the hazardous materials, so that specialists can be dispatched to the scene. Other warning signs include product names, company emblems, and other observable markings that describe the cargo (e.g., Exxon Gasoline); labels and markings on packages; distinctive vehicle or container configurations (especially tank trucks and tank cars); imme- diate notification by the truck driver or train crew of the hazard when reporting the incident; and the presence of smoke, fire, vapor, strong odor (e.g., gasoline fumes), or other sensory cues (e.g., eye and skin irritation). On-board documents, such as shipping papers, do not function as early warning devices, because they must be retrieved from the vehicle and are not readily observable. 1. 2. Basic description of the types of hazards present: Once the pres- ence of hazardous materials is recognized, responders need a descrip- TABLE 3-1 Responder Information Needs and Sources Critical Information Needs Possible Sources of Information Forewarning of the presence of hazardous materials Basic description of the general types of hazards present Technical names, quantities, and packaging of materials Diamond-shaped placards; prominent vehicle markings; vehicle container and configuration (e.g., tank); driver/train crew notification; smoke, fire, or strong odor Placards (symbols, coded colors, and class/division numbers); container labels, if visible; vehicle markings; driver/train crew notification Placards with UN/NA numbers; shipping paper entries; markings and labels on vehicle and packaging; DOT Emergency Response Guidebook; CHEMTREC; shipper's emergency telephone number
Necessity and Benefits of Improved Information 65 tion of the general types of hazards involvedâfor instance, whether the materials are explosive, flammable, poisonous, radioactive, or corro- sive. This information is necessary to allow first responders to begin isolating the area and to prepare to handle the incident. With this information, responders can consult the DOT Emergency Response Guidebook for preliminary guidance on general health and fire haz- ards, first aid procedures, and protective equipment. Placards are intended to be the principal indicator of hazard type, primarily through symbols keyed to the hazard class, coded colors, and hazard numbers displayed on the placard. Except for the Dangerous placard, all placards contain this information. For instance, the skull and crossbones indicates poison, an orange color indicates explosive, and the hazard number 8 (located at the bottom of the placard) indicates corrosive material. Other potential indicators of hazard type are descriptive company emblems and markings on the vehicle and labels on exposed packages (also, some bulk materials have the proper shipping name marked on the package). Whereas shipping papers provide this information, they are less effective than external sources. 3. Technical names, quantity, and packaging of materials: Once responders know the general type of hazard present, they can begin to handle the incident. But the preliminary information is seldom suffi- cient beyond the earliest stages of the response, after which more detailed information is essential. Responders need to know the techni- cal names of the materials involved, their quantities, and packaging. This information can be obtained through several sources accompany- ing the vehicle, including United Nations/North American (UN/NA) material identification numbers on placards and required entries on shipping papers, which give the product's technical name and the shipper's emergency telephone number. Only after the material is identified can responders seek more de- tailed information about its physical properties (e.g., flash point), incompatibilities with other materials, extinguishing agents, protective equipment, safe isolation and evacuation distances, and other product- specific information required to contain and mitigate the incident. Responders may call the shipper or CHEMTREC for this information; some of which is provided in the product's Material Safety Data Sheet (MSDS). In some larger communities, fire and police departments have access to computer software and other reference materials that contain handling information on thousands of hazardous substances.8 Haz- ardous materials experts and state and federal agencies also may be contacted for advice on mitigation procedures and on-scene assistance.
66 HAZARDOUS MATERIALS SHIPMENT INFORMATION If material identification is delayed or initially incorrect, the incident may be handled much less efficiently and potentially less safely. If material-specific information is not found, responders may continue to follow the general instructions provided in the DOT guidebook. The guidebook instructions are general in nature and cannot substitute for specific product information. Under these circumstances, responders may act with extreme caution as instructed in the guidebook, which may result in unnecessary evacuations, traffic delays, and inefficient use of response resources. Alternatively, responders who are not trained to handle hazardous materials incidentsâespecially when critical infor- mation is missingâmay act less conservatively, exacerbating the inci- dent and endangering safety. KINDS OF INFORMATION PROBLEMS ENCOUNTERED BY RESPONDERS What are the problems encountered by responders attempting to ob- tain information through existing sources, such as placards, shipping papers, and package labels? For each incident in the case studies, determinations were made on the basis of available evidence about the accessibility, timeliness, and reliability of the three critical information items discussed in the pre- ceding section. Cases were classified as having information problems when one or more were inaccurate, unavailable, not found, or not properly consulted by responders.' Inmost of the cases (101 of 125), the information sources available at the scene appear to have met the critical information needs of re- sponders in a timely and reliable manner. For example, correct placards were displayed and recognizable, shipping papers were accessible and complete, and the operator of the vehicle was helpful in identifying the material or providing responders with required documentation. In many cases the information provided appears to have been sufficient, so responders did not need to consult all available information sources, such as the shipper's emergency telephone number. Despite the successes, the information system appears to have failed regularly. Twenty-four cases had information problems, including 20 percent of the 40 California truck cases, 28 percent of the 40 NTSB truck cases, and 11 percent of the 45 NTSB rail cases (Table 3-2).
Necessity and Benefits of Improved Information 67 TABLE 3-2 Cases with Information Problems No. of No. with Information Source Cases Studied Problems California Truck 40 8 NTSB Truck 40 11 NTSB Rail 45 5 Altogether, about 25 percent (19/80) of the truck incidents had information problems, compared with roughly 10 percent of the rail cases. Many of the problems encountered were similar in nature, falling into six broad problem categories given in Table 3-3. Each category is described in this section, and illustrations of specific kinds of problems are given. The problem categories are not mutually exclusive, since many of the cases consisted of more than one type of problem. Summa- ries of the case studies with information problems, as well as the supplemental incidents, are provided in Appendix B. 1. Required sources of information were missing or inaccurate: The responders interviewed expressed concern about the frequency of miss- ing or incorrect placards and shipping papers at hazardous materials' incidents. On the basis of an analysis provided by the Office of Motor Carriers of the Federal Highway Administration (FHWA), state police data on truck inspections collected by FHWA under the Motor Carrier Safety Assistance Program (MCSAP) indicate that violations of federal requirements for placarding occur in about 30 percent of hazardous materials trucks inspected, whereas violations of shipping paper re- quirements occur in about 25 percent. About one in three trucks ticketed for placard violations and one in five trucks ticketed for ship- ping paper violations are put out of service by state inspectors because of the severity of the infraction. Although similar nationwide data are not available for railroads, state inspection data from Illinois, which has heavy rail traffic, indicate that placards and shipping papers are missing or incorrect in about 2 to 3 percent of rail shipments (Illinois Commerce Commission 1992; Illinois Commerce Commission, per- sonal communication). Missing or inaccurate placards, shipping papers, and other emer- gency information sources created problems for responders in 10 of the
TABLE 3-3 Information Problems in Truck and Rail Cases illustrations from Truck Illustrations from Rail Problem Category Specific Problems Cases (Appendix B) Cases (Appendix B) Required sources of information are missing or inaccurate Placards, shipping papers, or other information sources are obscured, destroyed, or inaccessible because of a crash or fire or threat of fire, explosion, or toxic exposure Placards missing or incorrect Shipping papers missing, incorrect, or incomplete Placards destroyed or obscured Shipping papers destroyed or inaccessible NTSB Case 9âtruck was not placarded NTSB Case 11âUN/NA number missing from orange panel NTSB Case 15âtruck placarded incorrectly California Case 18âillegal shipment of hazardous waste; truck was not placarded NTSB Case 7 and Supplemental Incident 7âplacards burned in fire Supplemental Incident 5â placards were not visible because of smoke and fire NTSB Case 2âshipper's name and telephone number missing from waybills NTSB Case 31âshipping papers listed wrong material and did not identify shipper NTSB Case 41âhazardous materials information in consist and waybills differed Supplemental Incident 3â shipping papers were located in front end of train, which was not accessible to responders because of wreckage and material hazards
NTSB Cases 1 and 7 and Supplemental Incidents 7 and 8âshipping paj,ers not retrievable because of wreckage and spill Supplemental Incident 8â placards were obstructed in wreckage Information sources are in compliance with regulations and accessible, but they fail to efficiently convey important information to responders Dangerous placard not descriptive of hazards Multiple shipping papers that are cumbersome to review Responders did not trust all information sources available and therefore did not use them California Cases 17 and 33âshipping papers were difficult to review because of varying formats from multiple shippers Supplemental Incidents 1 and 2âresponse delayed because of fire fighter concerns about potential inaccuracies in papers due to multiple shippers and greater chance of error NTSB Case 45 and Supplemental Incident 2âresponders could not immediately determine which cars were involved due to fire at accident site Supplemental Incident 3â responders could not match ejected packages with intermodal containers or derailed flatbed cars because of the scattered wreckage NTSB Case 15âcars carrying hazardous materials were not listed in consist (continued on next page)
TABLE 3-3 (continued) illustrations from Truck illustrations from Rail Problem Category Specific Problems Cases (Appendix B) Cases (Appendix B) Information is insufficient because the material or shipment is exempt from some federal hazardous materials transportation regulations Some potentially harmful materials or shipments are not covered by DOT regulations governing placarding, labeling, and shipping papers because they are not designated as hazardous, do not meet minimum weight/ quantity thresholds, or are not involved in a clearly transport-related activity Responders sometimes cannot readily confirm that a material is not hazardous California Case 3 and NTSB Cases 14 and 35â material not designated as hazardous by DOT; therefore warning placards were not displayed and shipping papers did not contain emergency information NTSB Case 4âmaterial not designated as hazardous by DOT because of low concentration of material; therefore placards were not displayed Supplemental Incident 3â truck not placarded because 650-lb shipment of hazardous material did not meet 1,000-lb placarding threshold California Case 7â semitrailer containing Supplemental Incident 3â responders were unable to determine proper response to pesticide that spilled from a tank car into a river. Placards and other emergency information did not accompany shipment because the material was not regulated in transport by DOT
Truck driver unable to identify basic hazards of cargo Train crew did not take proper actions to assist responders, such as identifying themselves and providing on-board documentation hazardous materials caught fire but was not placarded because it was used for storage NTSB Case 4âshipping papers in tank truck containing nonhazardous materials were destroyed. Responders could not determine contents and were concerned that the cargo might be hazardous NTSB Case 1âdriver misidentified material concentration to responders NTSB Case 19âdriver misidentified material to responders, causing confusion and an unnecessary evacuation Supplemental Incident 6â driver identified the material as radioactive but did not know the degree of radioactivity Vehicle operator is unprepared to provide information NTSB Case 45âtrain crew left accident scene with shipping papers Supplemental Incident 4â train crew left waybills and emergency response instructions in the train 0- CD 0 (continued on next page).
TABLE 3-3 (continued) Problem Category Responders fail to obtain or use available information Specific Problems Responders unaware of all information sources available or unprepared to take advantage of all available information Illustrations from Truck Cases (Appendix B) California Case 17â responders did not attempt to contact the shipper for advice on handling the incident for several hours, even though the shipper's emergency telephone number was available Illustrations from Rail Cases (Appendix B) NTSB Case 2âlocal fire officials did not know carrier's emergency telephone number and did not know the key elements of information that are needed by CHEMTREC to assist with response NTSB Case 41âresponders did not attempt to contact railroad immediately and did not attempt to contact CHEMTREC for product information and communication with shippers Nom: The six problem categories cover all problems identified in the case studies. Eight of the California truck cases had information problems, as did 11 NTSB truck cases and S NTSB rail cases. Specific problems in the cases are cited as illustrations of problems. For additional illustrations of problem types, 11 supplemental incidents (derived from newspaper and NTSB reports of major incidents) are also cited. Some cases are cited in more than one problem category. For some problem categories, only a few of the most representative cases are cited.
Necessity and Benefits of Improved Information 73 case studies. As discussed earlier, placards provide information that is essential in preventing first responders from unknowingly entering a hazardous situation. The NTSB truck cases provide examples of incidents with missing or incorrect placards. In one case, a tractor- semitrailer carrying industrial batteries containing corrosives was not placarded as required; hence, fire fighters were unaware of the corro- sive shipment until more than 1 hr after they had started extinguishing the fire near the hazardous cargo. In another case, placards were in place, but the required UN/NA number was missing, forcing fire fight- ers to retrieve the shipping papers from the truck to identify the bulk material. A problem in the case studies more common than missing placards was incorrect or incomplete shipping papers. In one of the NTSB rail cases, neither the shipper's name nor the emergency telephone number was entered on the shipment's waybill as required, resulting in a S-hr delay before emergency responders could identify the material. In several of the truck cases, shipping papers contained incorrect hazard class information (e.g., flammable, corrosive, or poison gas) or incom- plete descriptions of the quantity and packaging of the materials. In one of the NTSB truck cases, a Dangerous placard was displayed on a mixed truckload of corrosive and flammable shipments as allowed by regulation. However, the shipping papers did not specify the hazard class of each shipment (corrosive or flammable) as required. Re- sponders thus had to contact the shipper to obtain basic emergency information. 2. Information sources were obscured, destroyed, or inaccessible because of fire, wreckage, or other impediments: Because most critical information sources accompany the vehicle, they may be destroyed or become obscured or inaccessible during an incident because of fire, smoke, wreckage, or other impediments (e.g., presence of toxic vapor). Partly because of the small proportion of incidents with fires and severe wreckage, this problem occurred in only two case studies. In one NTSB truck case, the placards were burned beyond recogni- tion before responders arrived. The driver was killed, and no informa- tion could be found to identify the hazardous cargo. Fire fighters closed the highway and evacuated the area for several hours to let the vehicle burn instead of attempting to extinguish the fire without knowing the possible hazards. In another case the shipping papers of a truck carry- ing hazardous cargo burned in the cab. Although responders could
74 HAZARDOUS MATERIALS SHIPMENT INFORMATION recognize the placards, they could not take the most effective and expedient actions to mitigate the incident without the product-specific information contained in the shipping papers. Several of the supplemental incidents also illustrate how damaged or obscured placards and other information sources can lead to response difficulties with potential adverse consequences. In two rail incidents, both involving multiple-car derailments, placards were obscured by fire and wreckage. In the first incident, the train crew provided responders with the waybills, and therefore responders knew that hazardous materials were being carried on the train. Nevertheless, responders could not identify which cars contained hazardous mate- rials because the placards and other identification numbers were hid- den by smoke and debris. In the second incident, smoke and wreckage obscured the placards on several derailed tank cars, and two police officers who were first at the scene and unaware of the hazards entered the incident site unprotected (they were apparently uninjured). Sim- ilarly, in one of the supplemental truck incidents, which involved an overturned tractor-trailer carrying drums of poison, first responders (state police) were not aware of the hazardous cargo. The placards were not readily visible because of the orientation of the overturned vehicle. Several responders approached the vehicle and inhaled toxic fumes before a partially obscured placard was noticed at the front of the trailer. The supplemental incidents also demonstrate how shipping papers can be difficult to access during some incidents. One of the rail inci- dents involved a multiple-car derailment in which special hazardous materials response units could not approach the front of the train because the access roads were obstructed by wreckage and spilled cargo. As a result, they could not immediately retrieve the shipping papers, which were located at the front of the train. Eventually, after a lengthy delay and confusion, the papers were retrieved by helicopter. In the case of the overturned truck cited earlier, responders could not access the shipping papers for more detailed information on the poison because toxic fumes impeded entry into the vehicle's cab, where the papers were located. 3. Information sources were available and in compliance with regu- lations, but they failed to fully or efficiently convey essential informa- tion: The failure of available information sources to convey essential information to responders was more common in the case studies than destroyed or inaccessible placards and shipping papers, occurring in
Necessity and Benefits of Improved Information 75 eight of the case studies. In particular, the Dangerous placard, which warns of the presence of hazardous materials but does not identify the hazard type, was often a source of information problems. As discussed in Chapter 2, a vehicle or container may be transported under the Dangerous placard if it contains two or more shipments of hazardous materials that are in separate hazard classes as long as none of the individual shipments exceeds 5,000 lb. To determine the general haz- ards of a shipment placarded Dangerous, responders must search for information from shipping papers or on package labels and markings. Because many small shipments of hazardous materials are placarded in this manner, the problem is more common in the truck cases. Two of the California truck cases involved lengthy delays in responses, which resulted in large part from responders' confusion and uncertainty about the hazards of small shipments carried in vehicles bearing the Dangerous placard. In mixed loads of small shipments, many shipping papers may need to be reviewed by responders. If more than one shipper has provided papers, the formats may differ. Reviewing multiple shipping papers with varying formats can be cumbersome and time-consuming and can reduce responder confidence in the accuracy of the information. To confirm information in the papers or obtain more detailed information, it may be necessary for responders to contact individual shippers using the emergency telephone numbers provided in the shipping papers. Several such instances were found in the California truck cases. In one of the supplemental truck incidents, responders had to contact nine shippers to verify information on the shipping papers. Rail incidents in which several cars are involved in a crash or derail- ment can present similar problems. For instance, knowledge of the position of an individual car in the train is often necessary to quickly match the information in the shipping papers (train consist and way- bills) with placard and marking information on the cars. In severe wrecks or fires, responders may not be able to cross-reference the information sources, causing delays in shipment identification. In one of the supplemental rail cases, the identification problems were com- pounded because several intermodal containers and the drums carried within them were ejected from the flatcars on which they were carried. Responders had to rely on a combination of consist, waybill, placard, and package labeling information to match the drums with their con- tainers and the containers with their cars.
76 HAZARDOUS MATERIALS SHIPMENT INFORMATION In another NTSB rail case, the train consist did not accurately list the position of all the cars carrying hazardous materials, because several cars had been added to the train at a switching station. Federal regula- tions do not require an updated consist under these circumstances if the current crew members (i.e., the crew that performed the switch) have the correct waybills in their possession. In this case, the train crew provided responders with the correct waybills as required, but the discrepancies in the consist and waybills confused responders (who were not familiar with the regulatory allowances) and resulted in a slower response. 4. Essential information was not provided because the shipment was not covered by DOT regulations: In eight of the case studies, the shipments were considered dangerous by responders but were not covered by DOT regulations, resulting in little or no emergency infor- mation being available at the incident scene. This general type of problem occurred under several circumstances, including incidents in which the material was (a) not designated as hazardous by DOT (although it may be designated as hazardous by other regulatory agencies for nontransport situations), (b) shipped in quantities below the weight limit set by DOT for placarding, or (c) not involved in an activity that was clearly transport related and under DOT jurisdiction. Examples of the problem were also found in the supplemental inci- dents. One supplemental rail incident involved a herbicide (unregu- lated by DOT) that was released into a river following a tank car derailment. Because the material was not designated as hazardous by DOT, emergency information was not available, and responders were unsure of the hazards for several hours. In another supplemental case involving the crash of an unplacarded truck, responders were con- cerned about a 650-lb shipment of hazardous material that was not placarded because it was under the 1,000-lb weight limit set by DOT for placarding of Table 2 materials. In part because of their uncertainty about the material's hazards, fire officials closed the highway and evacuated nearby businesses for several hours until the material was identified and the incident cleared. Responders sometimes cannot readily confirm that a material in a vehicle is not hazardous. Police and fire officials are trained to treat transport incidents in which they suspect hazardous materials are involved as involving them until they can confirm otherwise. As a result, incidents involving tank trucks or tank cars or the presence of fire, leaking cargo, or a strong odor and fumes are often handled, at
Necessity and Benefits of Improved Information 77 least initially, as hazardous materials incidents even if placards are not evident. In such cases, responders may encounter significant delays in identifying the materials and determining whether hazards are indeed present. In one NTSB case an unpiacarded tank truck rolled into a river, making the shipping papers inaccessible to responders. By contacting the truck's owner, responders were eventually able to confirm that the tank truck did not contain hazardous cargo, but the confirmation took several hours. Most responders interviewed for this study reported that they com- monly encounter problems confirming that suspicious cargoes are nonhazardous. A fire fighter interviewed from New York City esti- mated that about 10 percent of the responses by the city's special hazardous materials unit are made to incidents that are later found to have involved nonhazardous cargoes. 5. Vehicle operator did not assist emergency responders in obtaining essential information: Train and truck operators are not required to have specific knowledge of the hazardous cargoes they carry. Neverthe- less, in many of the case studies in which information was deemed sufficient and timely by emergency responders, the truck driver and train crew were key information sources. Truck drivers were identified as the most important information source by more than one-third of the responders interviewed in California truck case studies; they pro- vided shipping papers to responders, described the general hazards of materials, notified responders of the incident, and knew the name and emergency telephone number of the carrier or shipper. Truck drivers transporting bulk shipments tended to be the most helpful, because they often carried the same product regularly to and from the same locations. Conversely, in six of the cases with information problems, the truck driver or train crew provided little or no helpful information. In some of the more severe truck crashes, the driver was killed or seriously injured and therefore unavailable to responders. In several other cases the driver was uninjured but could not provide any hazard information or gave incorrect information to responders, which caused confusion and hindered efforts to identify the materials. In one of the supplemental cases involving a burning tractor- semitrailer (in which the placards and shipping papers were destroyed), the truck driver identified the cargo as radioactive but did not know the degree of radioactivity, causing responders to assume a worst caseâ that highly radioactive materials were on boardâand order an evacua-
78 HAZARDOUS MATERIALS SHIPMENT INFORMATION tion of 100 people from the area. After more than S hr, responders determined that the truck contained low-level radioactive material and canceled the evacuation. In two NTSB rail cases, both involving a derailment and fire, the train crew failed to present the shipping papers to responders. In one case the crew did not identify themselves to responders and left the incident scene with both sets of shipping papers. In the other, the crew left the waybills along with other relevant emergency information in the train, even though they had an opportunity to retrieve them earlier in the incident. 6. Responders did not properly use available information: Haz- ardous materials incidents are infrequent in some jurisdictions, and local response personnel may have little or no experience responding to them. As a result, responders may not take advantage of all available information sources when an incident occurs. In four of the case study incidents (two truck and two rail), responders failed to properly use all available information. In part, this problem may reflect shortcomings in the information sources available (for example, the required design and format of shipping papers and placards may have inherent flaws that hinder use by responders). However, it also reflects the level of emergency training and planning on the part of the response agencies. For example, in two similar California truck cases involving mixed loads of hazardous materials, responders were concerned about the potential for material incompatibilities. In one incident responders immediately called CHEMTREC to communicate with the shipper and obtain detailed product information, which was valuable for determining appropri- ate response procedures. Responders to the second incident did not try to contact CHEMTREC or the shipper for several hours even though the shipper's emergency telephone number was listed on the shipping papers. The oversight resulted in a protracted and more costly response. The shipper was contacted 7 hr after the incident began by a private hazardous materials contractor called to assist with the response. In two rail incidents responders also failed to take advantage of readily available information. In one NTSB case involving the derail- ment and explosion of several tank cars, local fire and police officials were at the incident site for nearly 1 hr before they tried to contact the railroad to obtain information about the hazardous materials. In an- other case involving a tank car derailment, local responders did not
Necessity and Benefits of Improved Information 79 know the emergency telephone number of the railroad that passed through its jurisdiction. Hence, responders could not quickly identify the shipper to obtain product-specific information. FREQUENCY OF INFORMATION PROBLEMS How often do information problems occur on a national basis? The case studies are most helpful in illustrating the kinds of informa- tion problems encountered by responders at hazardous materials inci- dents. Their use in inferring the frequency of information problems on a national basis is a much more tenuous application because of the relatively small number of cases examined and the limited depth and statistical validity of the investigations. Nevertheless, without such estimates it is difficult to understand the magnitude of the problem. The 125 cases were selected without prior knowledge of whether sources of information were adequate. In this respect the sample was random, and the occurrence of information problems in the cases may roughly indicate the occurrence of information problems in the total population of hazardous materials incidents. Because no other credible data are available, the committee developed gross estimates using the case studies, recognizing that the results are tentative. Uncertainties in the case study findings are compounded by uncer- tainties in the incident statistics needed to estimate the frequency of information problems on a national basis. As discussed previously, the RSPA data have serious shortcomings when used to estimate total truck incidents, because incidents involving intrastate and local truck opera- tors are not counted, whereas numerous minor spills at truck terminals are overrepresented. State incident data are more relevant because reports are submitted by public responders, but the data vary greatly by state in quality and coverage. Absent better data, the estimates devel- oped next represent the committee's best judgment. National Incident Estimates A source of hazardous materials incident data that can be used to estimate the total number of truck incidents nationally is California's system, CHMIRS, the largest and most comprehensive state incident
80 HAZARDOUS MATERIALS SHIPMENT INFORMATION data base. In 1989 (the most recent year for which aggregate statistics have been reported) CHMIRS received 705 reports of truck incidents from state and local police, fire, and other response agencies (Califor- nia Office of Emergency Services 1991). Approximately two-thirds of California jurisdictions reported some or all of their incidents to CHMIRS (California Office of Emergency Services, personal commu- nication). If it is assumed conservatively (that is, to reduce the possi- bility of underestimating the number of incidents) that the data base accounts for between one-third and two-thirds of truck incidents in the state, between 1,000 and 2,000 incidents per year occur in California. California accounts for about 9 percentâor roughly 1 in 1 0âof the nation's heavy-truck miles (FHWA 1991) If the same ratio holds, the total number of truck incidents nationally is roughly 10,000 to 20,000 per year. To check this estimate, incident statistics can be compared from other states that collect incident data and have considerable hazardous materials traffic. Illinois, New York, and New Jersey reported a com- bined total of 915 truck incidents in 1989 (Hu and Casey 1992; Illinois Environmental Protection Agency 1991). This total may be low be- cause of incomplete reporting." If about one-half of all incidents are reported, nearly 2,000 incidents are expected in the three states com- bined. Together, New York, New Jersey, and Illinois account for ap- proximately 16 percent of the nation's truck miles, or about one in six (Fl-IWA 1991). If the three states also account for about one in six truck incidents, the total number of incidents nationally is about 12,000, which is within the range developed from the California data. The RSPA data are more reliable as an indicator of total rail inci- dents, because most rail shipments are carried by interstate carriers, which are required to file RSPA incident reports. Another potentially reliable source of rail data is CHEMTREC. Most large railroads report hazardous materials incidents to CHEMTREC as a standard practice because of shipper and receiver requests that they do so. In 1991, 1,146 rail incidents were reported to RSPA, whereas CHEMTREC received 1,397 notifications of rail incidents (RSPA 1991; CHEMTREC, per- sonal communication).12 These figures are fairly consistent, suggesting that between 1,000 and 1,500 rail incidents occur each year, or roughly 1 for every 10 to 20 truck incidents. Forty-one rail incidents were reported to CHMIRS in 1989, which results in a comparable ratio of 1 for every 17 truck incidents (41 to 705).
Necessity and Benefits of Improved Information 81 Number of Incidents with Information Problems The foregoing incident estimates are rough, but they provide an indica- tion of the total population of hazardous materials incidents. The estimates, along with the results from the case studies, allow for crude estimates of the number of incidents with information problems na- tionally (Table 3-4). In the 80 truck incident case studies, nearly 25 percent had information problems. At this rate, responders would encounter information problems in 2,500 to 5,000 truck incidents per year. In the 45 rail cases, about 10 percent had information problems. At this rate, there would be 100 to iSO rail incidents per year with information problems. A portion of incidents with information problems will have conse- quential outcomes. As discussed previously, 1989 CHMIRS data indi- cate that about 1 in 25, or 4 percent, of reported incidents involved a reportable injury, fire, explosion, evacuation, major highway closing, or other severe consequence. If applied to the foregoing estimates, this TABLE 3-4 Frequency of Truck and Rail Problems Incidents with Information Truck Rail Annual number of hazardous materials 10,000-20,000 1,000-1,500 incidents Incidents with information problems' Percentage 25 10 Number 2,500-5,000 100-150 Number of incidents with consequential 400-800 40-60 outcomesb Consequential incidents with information problems" Percentage 25 10 Number 100-200 4-6 aBased on findings from the 125 case studies, in which about 25 percent of truck incidents (19/80) and 10 percent of rail incidents (5/45) had information problems. 'Incidents with consequential outcomes involved a fatality, injury, evacuation, or major highway or track closing. The consequences. may or may not have been attributable to the information problem. Approximately 1 in 25, or 4 percent, of the truck and rail incidents reported to CHMIRS had a consequential outcome. This figure is used to estimate the number of incidents occurring nationally that have consequential outcomes.
82 HAZARDOUS MATERIALS SHIPMENT INFORMATION ratio suggests that responders encounter information problems in 100 to 200 consequential truck incidents and about S consequential rail incidents each year. However, these consequences are often the immedi- ate result of a crash or derailment. In the cases with immediate conse- quences, it is unlikely that improvements in the emergency information provided to responders would have altered the outcome, though some might have been avoided through preventive measures to avert the cause of the incident itself (e.g., hazardous materials routing restric- tions or driver training). COSTS OF INFORMATION PROBLEMS What are the costs of information problems and, therefore, the poten- tial benefits of resolving them? The low number of deaths attributable to hazardous materials trans- portation incidents, as reported by RSPA, is often cited in support of the performance of the existing system of emergency information. For example, RSPA concluded that in 1989 and 1990, 16 deaths occurred from hazardous materials incidents, and that all 16 occurred almost immediately, so that improved emergency information would not have changed the outcome (letter from DOT to Chairman of the Transporta- tion and Public Works Committee, U.S. House of Representatives, May 1990). As discussed earlier, however, RSPA's estimates of the number of deaths in hazardous materials incidents may be incomplete, primarily because they do not include incidents involving intrastate and local carriers and because they rely on self-reporting by carriers. Even if the fatality data are complete, deaths are not the only potential cost of inadequate or untimely information at hazardous materials incidents. Others include avoidable injuries, traffic delays, evacuations, ineffi- cient use of response resources, environmental damages, and spill cleanup costs. In the following discussion, some of the major costs associated with information problems are considered. Some are quantified using sim- plified assumptions and the limited findings from the case studies. The resulting cost estimates in Table 3-5 are provided only to illustrate the magnitude of the problem and are not intended to be conclusive or complete.
Necessity and Benefits of Improved Information 83 TABLE 3-5 Potential Costs of Incidents with Information Problems Item4 Costb Injuries (number)c 300-600 Property loss ($ millions) 50-100 Major highway closings Number 40-80 Cost ($ millions) 20-40 Public evacuations Number 25-50 Cost ($ millions) 25-50 Response resources d 4Only items quantified in Chapter 3 are presented. bCosts are estimated to illustrate their possible scale. Only a portion of the costs are caused by information problems per Se. 'Potential long-term and delayed injuries and illnesses, which are not reported in incident data bases, are not included in the estimate. dlnformation failures, even if they occur in a minority of incidents, result in higher average response costs for all in- cidents because they degrade the efficiency of utilization of emergency response resources and increase the amount of response capacity needed. The total public-sector cost of response and preparation for all hazardous materials trans- portation incidents is estimated to be $125 million to $250 million annually. A substantial reduction in the frequency of information problems would reduce this cost by an unknown amount. Injuries to Emergency Response Personnel and Others Missing or inaccurate information could result in responders taking inappropriate or unsafe actions at hazardous materials incidents, possi- bly resulting in greater exposure and injury to emergency response personnel and the public. In the full set of 125 cases, all of which had consequential outcomes, there were 24 fatal and 236 nonfatal injuries, for a total of 260. In the 99 cases without information problems, there was an average of nearly two reported injuries per incident. In the 26
84 HAZARDOUS MATERIALS SHIPMENT INFORMATION cases with information problems there was an average of three reported injuries per incident, including one incident that resulted in 55 injuries. If the higher rate of injury occurred in the estimated 105 to 205 incidents with information problems, the result would be 300 to 600 injuries per year. It is unclear how many of these injuries might be related to informa- tion problems. The average number of injuries in the 26 cases with information problems did not differ significantly from the average in the 99 cases without apparent problems. Thirteen of the 26 cases with information problems resulted in injuries. The injuries occurred before the arrival of emergency responders in 11 of the 13. It is not evident that better information would have resulted in fewer injuries in these cases. The injuries occurred almost immediately, usually during the crash, and were typically unrelated to the hazardous cargo. In the two cases in which injuries occurred after the arrival of emergency responders, the injuries were caused by smoke inhalation and heat stress incurred by police and fire fighters. Although some of these injuries may have been related to information problems, that could not be determined from available evidence. Injury statistics can be misleading for several reasons, in particular because of inconsistent and incomplete reporting. For instance, in some cases all persons requiring a hospital visit, even for observation pur- poses, were reported as injured; in other cases only persons admitted to the hospital were reported as injured. During the case study investiga- tions, the committee counted all kinds of injuries to the extent possible. Nevertheless, some injuries, such as delayed injuries or long-term ill- nesses and other adverse health effects, are not reported in incident data bases or other documentation used to develop the case studies. Few incidents receive the thorough evaluation and follow-up required to identify possible long-term effects. As a result, injury counts derived from the case studies may be incomplete, and some incidents that resulted in chronic or delayed health effects may not have been reported in the case studies and national estimates of consequential incidents. The case studies indicate that cautious behavior by responders can reduce the risk of injury. Inmost cases the greatest chance of injury is at the outset, if the presence of hazardous materials is not recognized by first responders. If placards and other prominent warning information are missing or obstructed, first responders may not take sufficient precautions and may approach the incident site unprotected. This
Necessity and Benefits of Improved Information 85 occurred in one of the supplemental incidents cited earlier, in which a first responder inhaled toxic fumes from an overturned truck with obscured placards. Also, responders may not be aware of the need to immediately evacuate nearby residents or workers, thereby exposing them to the hazards. Most information problems in the case studies involved material identification after the presence of hazardous materials was known or suspected by responders rather than missing or obscured placards or other warning information. Under these circumstances, the responders tended to act cautiously until the material was confidently identified. They were often willing to accept the costs associated with possibly unnecessary evacuations, highway closings, and other precautions as a trade-off for greater assurance of safety. The recurrence of this cautious behavior by responders in the case studies suggests that the chances of injury are lowered most significantly when responders are initially warned of the presence of hazardous materials, even when more de- tailed hazard information is not available immediately. On the other hand, the safest course of action may not always be taken by re- sponders, especially if training for hazardous materials incidentsâ including instructions for handling incidents with information prob- lemsâhas been lax. Property Losses Missing or untimely response information can increase property dam- age by causing responders to handle the incident incorrectly or, more likely, to act more cautiously. For instance, they may allow a vehicle and its contents to burn rather than risk the safety of response personnel when the hazards are unknown. DOT collects data on property losses from incident reports filed by interstate carriers. The data are likely to be incomplete because carriers are not aware of all the property losses associated with an incident, such as public-sector costs (e.g., expense of road repair). Some case study incidents resulted in relatively minor property damage because no wreck or fire occurred. Others resulted in major fires and crashes that destroyed vehicles, their cargoes, and surrounding property (for instance, in one case several evacuated homes were burned). Property loss information was not obtained for the case studies, but it is likely that losses totaled in the millions of dollars in many of the severe cases.
86 - HAZARDOUS MATERIALS SHIPMENT INFORMATION The median property loss in the 45 NTSB rail cases, which had the most complete cost information, was $500,000, and 13 of the 45 cases had losses exceeding $1,000,000. To estimate the magnitude of this cost, one can assume average property damage of about $500,000 per consequential incident to cover damage to vehicles, cargoes, and surrounding property, including road and track facilities. If such property losses were to occur in the estimated 105 to 205 consequential truck and rail incidents with information problems, total losses would range from $50 million to $100 million. It is unclear what share of this cost might be attributable to information problems and therefore potentially avoided by informa- tion improvements. Highway Closings and Delays Many responders who recognize or suspect that hazardous materials are present but are unable to obtain critical information on the material will act conservatively. This may result in evacuations, highway and track closings, and other extreme precautions (such as shutting down water supplies). The purpose of these actions is to ensure the safety of the public and response personnel until better information can be obtained. Costly precautions might be avoided if better information were available earlier. Highway closings and traffic diversions are among the most common precautions taken by responders. Because of the nature of hazardous materials incidents, many highway closings and prolonged traffic de- lays cannot be avoided. In some situations responders may close a highway as a safeguard until credible information on the hazardous cargo is obtained. If a major urban highway or commuter route is affected, the traffic delays and associated costs can be substantial. Glickman has estimated that three major hazardous materials truck incidents in the Washington, D.C., area resulted in traffic delay costs exceeding $4 million in each case, including the value of motorists' time and extra fuel consumption (Glickman 1991). Of the 80 truck cases examined, 23, or about 28 percent, involved closings of Interstate or other primary highways, averaging 10 hr each. Altogether, about 40 percent of the truck cases with information prob- lems (8 of 19) resulted in major highway closings compared with about 25 percent of the truck cases without problems (15 of 61). If the 40
Necessity and Benefits of Improved Information 87 percent rate of closings were to occur in the estimated 100 to 200 consequential truck incidents with information problems, there would be 40 to 80 major closings. The roughly one-third lower rate of closings found in the case studies without information problems suggests that many of these closings might be avoided by information improvements. Glickman's estimate of more than $4 million per major highway closing is probably high because it was developed for incidents that occurred in a major metropolitan area on highways with high traffic volumes. Many highway closings occur in rural areas. Even in these cases, the cost may be significant because the roads most likely to be closed are Interstate or other major highways that are principal routes of hazardous materials movement. If the cost per incident is only 10 percent of the cost developed by Glickman, or about $500,000 per closing, the total cost for 40 to 80 highway closings would be $20 million to $40 million. How much of the cost to attribute to informa- tion problems is uncertain. In the case studies about one-third fewer closings (on a per incident basis) occurred in cases without information problems. Evacuations Evacuations occur in a small percentage of hazardous materials inci- dents but are not rare, especially in major incidents. On one hand, poor information may result in unnecessary evacuations as responders as- sume the worst to ensure public safety. On the other hand, poor information could result in responders not knowing the full extent of the danger and not ordering an advisable evacuation. Most evacuations ordered during hazardous materials incidents oc- cur at workplaces, especially truck terminals and shipper loading docks where many shipments of hazardous materials are handled. These evacuations tend to involve fewer than 200 workers and are usually short. Public, or residential, evacuations are less common but tend to be longer and entail more significant costs and inconveniences to a larger number of people. Of the 26 truck and rail cases with information problems, 6, or about 25 percent, involved evacuations of 200 or more people. In the 99 case studies that did not have information problems, 15, or about 15 percent, resulted in public evacuations. If the experi- ence in the case studies is representativeâthat is, about 25 percent of consequential incidents with information problems result in public
88 HAZARDOUS MATERIALS SHIPMENT INFORMATION evacuations-25 to SO public evacuations could be expected among the estimated 105 to 205 consequential truck and rail incidents with information problems. The costs associated with public evacuations can be difficult to quantify. They include inconvenience to the public, lost productivity by businesses, and dedication of resources by the govermnent agencies and other organizations assisting with the effort. The following simplified estimate illustrates the potential magnitude. In the case studies, public evacuations averaged 1,000 people. If the average cost per evacuee is $1,000âaccounting for direct evacuation costs and lost wages and earnings by residents and businessesâthe total cost for each evacuation is $1 million." The cost of 25 to SO evacuations would be $25 million to $50 million. It is unclear what share of these costs is related to information problems. The higher rate of evacuations in the case studies with information problems (25 percent compared with 15 percent of cases without information problems) suggests that about one-third of the estimated 25 to SO public evacuations might be avoided by better information. The severity of evacuations differs from incident to incident, some- times substantially. For instance, in the 125 case studies, the largest evacuation consisted of 4,000 people, whereas 30,000 people were evacuated in one of the supplemental rail incidents. Such major evacua- tions occur infrequently, but when they are required, the associated costs are undoubtedly far higher than the estimated $1 million per incident. Inefficient Use of Response Resources Major hazardous materials incidents can involve hundreds of police, fire fighters, emergency medical personnel, and other responders. Hence, even minor information problems that result in protracted responses can be costly. Information problems that result in prolonged or unnecessary responses at smaller, more routine incidents involving fewer responders are not nearly as costly on an individual basis. How- ever, the aggregate cost of thousands of these incidents is surely much higher. For most hazardous materials incidentsâwhich seldom involve injuries, evacuations, or major highway closingsâthe most significant cost is the time of hazardous materials units and other response person-
Necessity and Benefits of Improved Information 89 nel and the equipment and supplies used. Earlier in this chapter it was estimated that emergency responders encounter information problems in 2,500 to 5,000 truck incidents and 100 to 150 rail incidents each year, including both consequential and nonconsequential incidents. It is difficult to determine the effect of information problems on state and local spending on emergency response. One effect may be that communities increase their response capacity to compensate for the loss in efficiency in the use of emergency response resources. For instance, if fire personnel and equipment are delayed at an incident because of poor information, they cannot respond to other incidents occurring at the same time. Delays may require communities to in- crease the size of their emergency response forces to maintain an acceptable response capability. In cases where responders cannot identify the material, they may use special protective clothing, extin- guishing agents, and other mitigation equipment and supplies that might not have been used if the material had been identified quickly. Some of these articles are expensive and cannot be reused. For instance, some protective outfits designed for a single use can cost several hundred dollars to more than $1,000. In small communities with limited budgets for hazardous materials responses, one such incident can result in the exhaustion of years' worth of expensive hazardous materials supplies. Currently, state and local fire departments across the country spend approximately $500 million per year to respond to and prepare for hazardous materials incidents in transportation and at fixed facilities.14 Hazardous materials incident data collected by California and Illinois (California Office of Emergency Services 1991; Illinois Environmental Protection Agency 1991) indicate that between 25 and SO percent of all hazardous materials incidents occur in transportation or at transporta- tion facilities, suggesting that between $125 million and $250 million may be related to transportation incidents. If information problems occur in 10 to 25 percent of these incidents, as in the truck case studies, they are likely to contribute to the $125 million to $250 million cost. For instance, information failures, even if they occur in a minority of incidents, are likely to result in greater general mistrust of all informa- tion by police and fire fighters responding to hazardous materials incidents. Consequently, information failures can result in slower and more costly responses to virtually all kinds of hazardous materials transportation incidents, including many in which information is credible.
90 HAZARDOUS MATERIALS SHIPMENT INFORMA11ON Other Costs Other types of costs that have not been discussed may stem from poor information. 'Whereas they are potentially as significant as the costs discussed earlier, they are more difficult to characterize and quantify, even in illustrative terms. Examples include the following: Environmental damage and cleanup: Poor emergency response information that results in a slow, inappropriate, or overly cautious response could result in greater pollution and other environmental damage due to a release of hazardous substances into the air, ground, or water. In these instances, information problems can lead to higher cleanup and remediation costs, including those associated with spill removal (e.g., from soil and water), clearance of contaminated wreckage, and repair of damaged tracks and roadways. The railroad industry alone spent an estimated $100 million on cleanup and remedi- ation of major hazardous materials spills from 1980 to 1989 (Barkan et al. 1991). Environmental damage and cleanup costs are seldom accurately recorded in incident data bases. The environmental effects of haz- ardous materials spills are increasingly becoming an important consid- eration of industry, especially because of liability concerns and cleanup costs. In recent years, some efforts have been made (especially by railroads) to monitor these effects, although available data are insuf- ficient to estimate costs. Yet in most cases the overriding concern of emergency responders is protecting lives. Therefore, reductions in envi- ronmental damage and cleanup costs resulting from information im- provements are most likely to occur as a by-product of better response measures to ensure safety. Chronic health effects: Many fire fighters and other emergency responders are concerned about the long-term health effects of expo- sure to hazardous materials. Evidence of these effects cannot be found in incident injury data. Poor information that leads to an improper response can increase the chances of exposure of both the public and response personnel, but few data are available and no conclusive studies have been conducted on the subject to determine costs. Carrier and shipper productivity losses: Carriers and shippers have an interest in ensuring that incidents involving their vehicles and cargoes are handled efficiently by responders. Poor emergency informa-
Necessity and Benefits of Improved Information 91 tion that impedes the actions of emergency responders can cause idle- ness of transportation vehicles, lading, terminal facilities, and carrier and shipper personnel, thereby reducing productivity. The greatest productivity losses, at least collectively, are likely to result from delays at minor incidents that are prolonged by the search for hazardous materials information. An efficient response may allow the vehicle and driver to return to operation sooner. Because thousands of hazardous materials incidents occur each year, the associated delay costs to ship- pers and receivers may be significant in the aggregate. Effects on responses to all incidents: Information failures may have far-reaching effects on the practices of emergency responders and the use of emergency response resources. Information failures can reduce the confidence of responders when making decisions at the scene of an incident. This could produce slower and less appropriate responses to all hazardous materials incidents and in the long run lead to greater expenditures on emergency response capacity and more unnecessary traffic delays, evacuations, and property losses. The reduction of sec- ondary effects and their associated costs could be significant benefits of ameliorating information problems. Catastrophic Incidents None of the cost items discussed earlier includes costs associated with catastrophic incidents. Although the case studies consisted of consequential incidents, they did not include incidents that involved numerous fatalities. Catastrophic events are difficult to analyze statis- tically because they are rare and therefore seldom found in historical records of incidents. Nevertheless, because of their consequences, they are of great concern to responders and the public. Statistics collected by Glickman indicate that 10 hazardous materials truck and rail incidents with 10 or more deaths occurred in the United States between 1945 and 1989, and 17 others resulted in more than S deaths (Table 3-6). Six of the 27 incidents were caused by fuel trucks being struck by trains at highway grade crossings. In all of the cases with 10 or more deaths, the materials involved were common petro- leum products (home heating oil, gasoline, butane, and propane).15 In 6 of the 10 cases, the deaths were caused by explosions or fires occur-
TABLE 3-6 Hazardous Materials Truck and Rail Incidents that Caused Five or More Deaths, 1945 to 1989 (Glickman et al. 1992) Year Location Mode Hazard Type Release Type Number of Deaths4 1950 Chicago, Ill. Truck Flammable liquid Fire 34 1959 Monroe, La. Rail Flammable gas/liq. Fire 8 1959 Deer Park, Pa. Truck Flammable gas Explosion 11 1959 Meldrim, Ga. Rail Flammable gas Explosion 23 1960 Bakersfield, Calif. Truck" Flammable liquid Explosion 14 1962 New Berlin, N.Y. Truck Flammable gas Explosion 10 1966 Everett, Mass. Truck" Flammable liquid Fire 12 1969 Crete, Nebr. Rail Nonflammable gas Toxic vapor 9 1969 Flint, Mich. Truck Flammable liquid Explosion 6 1970 Hudson, Ohio Truck Flammable gas Explosion 6 1972 Lynchburg, Va. Truck Flammable gas Explosion 6 1973 Kingman, Ariz. Rail Flammable gas Explosion 13 1974 Climax, Tex. Rail Flammable gas Explosion 7
1974 Decatur, Ill. Rail Flammable gas Explosion 7 1975 Eagle Pass, Tex. Truck Flammable gas Explosion 17 1976 Houston, Tex. Truck Nonflammable gas Toxic vapor 6 1977 Beattyville, Ky. Truck' Flammable liquid Fire 7 1977 Galveston, Tex. Rail Flammable gas Explosion 18 1978 Waverly Hill, Tenn. Rail Flammable gas Explosion 25 1978 Youngstown, Fla. Rail Nonflammable gas Toxic vapor 8 1980 Kenner, La. Truckb Flammable liquid Fire 7 1980 New Orleans, La. Truckt' Flammable liquid Unknown 6 1981 Huntsville, Ala. Truckt Flammable liquid Unknown 7 1981 Canon City, Cob. Truck Flammable liquid Unknown 8 1982 Oakland, Calif. Truck Flammable liquid Explosion 7 1988 Kansas City, Mo. Truck' Other Explosion 6 1988 Memphis, Tenn. Truck Flammable gas Fire 9 Maximum reported. tTruck carrying hazardous material struck by train at a highway-railway crossing. 'Incident involved a semitrailer used at a construction site to store blasting agents. Attempting to control a fire at the site, six fire fighters were killed when the unplacarded trailer exploded.
94 HAZARDOUS MATERIALS SHIPMENT INFORMATION ring instantaneously or within minutes of the incident, before emer- gency responders had arrived at the scene. In the other four incidents, explosions occurred during emergency response operations, killing fire fighters, other responders, and bystanders. In all four cases, however, the hazardous materials were already known to emergency responders; the explosions occurred because of faulty equipment and poor han- dling procedures.16 Past incidents are not sufficient to determine whether information problems will contribute to future catastrophes, because these events are both random and unusual. Hence, the potential costs of cata- strophic incidents that might be avoided by information improvements are not predicted in this study. However, one of the fatal incidentsâa 1988 incident in Kansas City, Missouri, in which six fire fighters were killedâis an example of a case in which better information might have saved lives. In the Kansas City case, the deaths were caused by an explosion of blasting agents stored in a trailer set afire by an arsonist at a construc- tion site. The trailer did not bear DOT placards because it was not in use in transportation and therefore not subject to DOT regulation. In fact, other federal regulations then in effect, under the authority of the Bureau of Alcohol, Tobacco, and Firearms, prohibited the display of DOT placards on vehicles storing explosives subsequent to transporta- tion. An investigation by the National Fire Protection Association concluded that the fire fighters were warned of the possibility of explosives en route to the incident and apparently learned at the scene that the burning trailer did indeed contain explosives. However, the investigation concluded that before the alarm, fire officials did not know the type, quantities, or locations of the blasting agents (National Fire Protection Association 1989). If such information had been avail- able at the time of the alarm through a central reporting system or other means, fire fighters might have been able to plan how to handle the incident safely. To have been useful, an information system would have had to cover hazardous materials in vehicles used for storage, which are not currently covered by DOT regulations (with the exception of en route shipments in temporary storage). The lack of emergency informa- tion in incidents that are not plainly transport related was identified as a problem in the case studies discussed earlier in the chapter. It illus- trates the need for coordination among the federal agencies responsible for regulation of hazardous materials.
Necessity and Benefits of Improved Information 95 SUMMARY Emergency responders to hazardous materials incidents must be pro- vided with critical information in a timely and reliable manner. They must be immediately warned that hazardous materials are present, know the general types of hazards involved, and quickly thereafter be able to identify materials, their quantities, and packaging. Few data are available to evaluate the performance of the existing information system. Hence, the committee examined 125 incidents, or case studies, to judge the system's performance. The case studies; supplemented with other incident examples and interviews with emer- gency responders, revealed the following six broad categories of infor- mation problems: Required sources of information, including placards and shipping papers, were missing or inaccurate; Placards, shipping papers, and other information sources were obscured, destroyed, or inaccessible because of wreckage, fire, or other impediments; Information sources were in compliance with regulations and were accessible but failed to fully or efficiently convey essential information; Information sources were not provided because the shipment was not covered by DOT regulations; S. The vehicle operator did not assist emergency responders in ob- taining necessary information; and 6. Responders did not properly use available information. Although the small number of case studies prevented a clear ranking of the frequency of problems, some categories recurred more often than others. A common problem was the failure of some required informa- tion sources to meet the information needs of responders. These cases often involved trucks carrying mixed loads of hazardous cargo. In this circumstance, placards do not always indicate the types of hazards present (e.g., the Dangerous placard), shipping papers may be cumber- some to review (because there may be several), and drivers are less likely to know the hazards of their cargo. On the other hand, few cases of destroyed or inaccessible placards and shipping papers were observed, partly because most hazardous materials incidents do not involve severe wreckage, fires, or explosions.
96 HAZARDOUS MAThRIALS SHIPMENT INFORMATION In most of the case study incidents the existing information system conveyed critical information to responders in a timely and accurate manner. Nevertheless, responders encountered information problems in about 25 percent of truck incidents and 10 percent of rail incidents. Results from the case studies suggest that responders encounter infor- mation problems in 2,500 to 5,000 truck and 100 to 150 rail incidents per year. Of these, about 100 to 200 truck and S rail incidents result in fatalities, injuries, evacuations, major highway closings, or other severe consequences. Information problems were not linked directly with injuries in any of the case studies. However, all types of injuries and other adverse health effects (e.g., long-term or delayed illness) may not have been reported, and other examples of incidents were found in which injuries resulted from poor information. The case studies indicate that responders often act cautiously when confronted with missing or unreliable information. This behavior may reduce the chances of injury and exposure in many cases, but evacua- tions, property losses, highway closings, and other costs may be in- creased as a trade-off. Rough estimates of these costs suggest that they total in the tens of millions of dollars per year, at a minimum. Some costsâsuch as environmental damageâcannot be quantified, even in rough terms, although better information may reduce them in many cases. Similarly, information improvements might increase the effi- ciency of emergency response to all incidents (by improving the confi- dence of responders) and reduce the consequences of rare catastrophic incidents. These benefits are also not quantifiable. The findings in this chapter indicate that information problems often result from fundamental shortcomings in training, regulations, and enforcement. The benefits of reducing these problems and their associ- ated costs total in the tens of millions of dollars per year for costs that can be quantified in gross terms. Whereas this estimate is undoubtedly incomplete, it suggests that moderate investments in information im- provements may be justified. NOTES 1. RSPA requires interstate carriers to report incidents in which there is (a) a release of hazardous materials; (b) a fire, spillage, or breakage of suspected contamination involving a shipment of radioactive or etiologic material; or (c) one of the following consequences that is a direct result of a hazardous material: fatality or injury requiring hospitalization; property damage exceed-
Necessity and Benefits of Improved Information 97 ing $50,000; public evacuation that lasts more than 1 hr; or closing of a major roadway, track, or other transportation facility more than 1 hr. Despite the predominance of a few carriers in reporting, the RSPA data are often used to present a historical picture of the safety experience of the entire industry and characterize hazardous materials incidents (e.g., incident loca- tions, consequences, and types of materials and containers involved). Annual fluctuations by a few carriers in their hazardous materials traffic and the diligence of their incident reporting have significant effects on year-to-year changes in reported incidents. Also, because LTL carriers carry many small shipments and operate many terminals, the data are skewed toward incidents occurring during loading operations and involving certain nonbulk materials and containers. Hence, the data are not representative and must be used with caution. One of the drawbacks of state data is that many states define spills from vehicular fuel tanks as reportable incidents. California, for instance, requires reporting of fuel tank spills of more than 40 gal. Because RSPA incident reports are submitted by carriers rather than emer- gency responders, it is improbable that the existing data base could provide these data reliably. S. Most of the derailments involved track closings; however, the severity and duration of these closings could not be determined on the basis of NTSB documentation. Information problems can be expected more often in severe incidents (e.g., a train derailment resulting in fire) because responders are under more stress and because of accompanying fires, wreckage, and other impediments to emer- gency information. To the extent that information problems contribute to severe consequences, a higher share of consequential incidents can be expected to have information problems. Fire fighters and other emergency responders from Baltimore, Chicago, Los Angeles, New York City, Phoenix, and Virginia were interviewed. Industry emergency handling experts from I.E. du Pont de Nemours and Company, Inc., Roadway Express, Inc., and Union Pacific Railroad Company were interviewed. Examples of other more detailed reference sources are CAMEO and the Association of American Railroad's manual, Emergency Handling of Haz- ardous Materials in Surface Transportation. These sources are intended to provide more product-specific information than the DOT Emergency Re- sponse Guidebook. Currently, a special committee of the U.S. Fire Administra- tion is studying the consistency of information in the guidebook and the other manuals and reference sources. An information source that is not properly consulted may be the result of poor design or insufficient training on the part of responders. Reliable data on hazardous materials truck mileage by state are not available. On the other hand, some of these incidents are reports of vehicular fuel tank spills.
98 HAZARDOUS MATERIALS SHIPMENT INFORMATION An annual average of 1,104 rail incidents was reported to RSPA from 1987 to 1991. Few studies have attempted to estimate costs associated with evacuations. After reviewing the literature, a study by the U.S. Nuclear Regulatory Com- mission calculated the average cost per person to be between $600 and $1,800 for those costs that could be quantified (Witzig and Shillen 1987). About $20 billion per year is spent by state and local fire agencies (U.S. Bureau of the Census 1988). According to statistics provided by several municipal fire departments, about 2.5 to 5 percent of their budgets are devoted to hazardous materials response and preparation (for both transportation and fixed-site incidents). This suggests that total spending on hazardous materials response and preparation is between $500 million and $1 billion per year. Newspaper accounts for each incident with 10 or more deaths were reviewed by the committee to determine incident circumstances. New York Times newspaper reports of each of the 10 incidents were reviewed to determine events causing fatalities. REFERENCES ABBREVIATIONS FHWA Federal Highway Administration NTSB National Transportation Safety Board OTA Office of Technology Assessment RSPA Research and Special Programs Administration Barkan, C., T. Glickman, and A. Harvey. 1991. Benefit-Cost Evaluation Using Different Specification Tank Cars To Reduce the Risk of Transporting Environ- mentally Sensitive Chemicals. In Transportation Research Record 1313, TRB, National Research Council, Washington, D.C., pp. 33-43. California Office of Emergency Services. 1991. Hazardous Material Incidents in Californiaâi 989. Hazardous Materials Division, Sacramento. FHWA. 1991. Highway Statistics 1990. U.S. Department of Transportation. FHWA. 1992. FY 1991 Annual Report of the Office of Motor Carrier Field Operations. U.S. Department of Transportation. Glickman, T. 1991. Highway Robbery: Social Cost of Hazardous Materials Inci- dents on the Capital Beltway. In Transportation Research Record 1313, TRB, National Research Council, Washington, D.C. Glickman, T., D. Golding, and E. Silverman. 1992. Acts of God and Acts of Man: Recent Trends in Natural Disasters and Major Industrial Accidents. Discussion Paper CRM 92-02. Center for Risk Management, Resources for the Future, Washington, D.C. Hu, P., and L. Casey. 1992. A Comparison of Hazardous Materials Release Incident Reports Submitted to DOTIRSPA's HMIS with Incident Reports Sub- mitted to the States of New York, New Jersey, and California. Report DHM-60.
Necessity and Benefits of Improved Information 99 Office of Hazardous Materials Planning and Analysis, RSPA, U.S. Department of Transportation. illinois Commerce Commission. 1992. Annual Report for 1991. Springfield. illinois Environmental Protection Agency. 1991. Annual Report on Hazardous Material Incidents in Illinois for 1990. Springfield. National Fire Protection Association. 1989. Fire Investigation Report: Blasting Agent Explosion; Six Fire Fighters Killed, Kansas City, Missouri, November 29, 1988. Quincy, Mass. NTSB. 1988. Safety Study: Case Studies of 189 Heavy Truck Accident Investiga- tions. Report NTSB/SS-8 8-OS. Washington, D.C. NTSB. 1991. Safety Study: Transport of Hazardous Materials by Rail. Report NTSB/SS-91-01. Washington, D.C. OTA. 1986. Transportation of Hazardous Materials. Report OTA-SET-304. Wash- ington, D.C. RSPA. 1991. Annual Report on Hazardous Materials Transportation: Calendar Year 1990. U.S. Department of Transportation. U.S. Bureau of the Census. 1988. Government Finances. Series GF. Witzig, W. F., and J. K. Shillen. 1987. Evaluation of Protective Action Risks. Report NUREGICR-4726. U.S. Nuclear Regulatory Commission.
Options To Improve Information: Technological Components The steps necessary to design and implement information technol- ogy applications that would give public-sector emergency responders access to information valuable for reducing losses from hazardous materials transportation incidents are described here. If government or industry set out to systematically design a system for this purpose, following the experience gained in development of other information systems and seeking to minimize development time, they would take three steps before undertaking full-scale implementation: 1.System requirements would be defined in terms of information needs and performance criteria. To establish system requirements for reliability, timeliness, coverage, and information content, the starting point would be analysis of past incidents and solicitation of emergency responder views to reach an understanding of responders' needs and the performance of the existing information system. Options for designs to meet the performance criteria would be specified. The specifications would cover technical aspects as well as the regulatory and administrative framework. Consideration of the poten- tial for making use of existing information capabilities to enhance cost- effectiveness and reliability would be an essential part of this step. The options would be introduced and evaluated in stages to determine whether they met the performance criteria, refine the de- 100
Options To Improve Information: Technological Components 101 signs, find the most cost-effective approaches, and determine whether costs would be justified by the benefits. Evaluation would rely on field testing of prototypes and on ongoing evaluation of initial implementa- tions. Decisions would be made in stages about whether to proceed further with implementation. These steps can be pursued to some extent in parallel. However, each step is necessary to avoid premature and ineffectiye attempts to intro- duce information technology in hazardous materials transportation emergency response. The steps were not followed in the development of the original central reporting system proposalâthe proposal was not founded on analysis of responder information needs, design options were not considered, and testing and prototype implementations were not conducted or proposed. Each of these steps is explained in the following sections. A limited start-up system is proposed in the final section as the most effective means to address design issues and begin evaluation. SYSTEM REQUIREMENTS Information Needs A thorough understanding of the information needs of emergency responders and the causes and costs of failures of the existing system in meeting those needs is essential for guiding the design of improvements and evaluating the benefits of improvements. The results of Chapter 3 allow a beginning of identification of information needs, an estimate of the level of reliability of the existing system (i.e., the frequency of information failures), identification of the six categories of causes of failures, and estimates of the benefits of improved information. These determinations are preliminary, because they are based on analysis of only 125 cases. More analysis of incidents is required to guide improvements. Performance Criteria Performance criteria are benchmarks with which the improved infor- mation system can be compared to judge whether it is producing a worthwhile improvement. In general, the more demanding the
102 HAZARDOUS MATERIALS SHIPMENT INFORMATION benchmarks, the more costly the system will be. Measurements would be made during testing, start-up, and operation of the improved system to determine whether the system was working properly. The primary criteria are as follows: an improved system should reduce the costs of hazardous materials transportation incidents, espe- cially deaths and injuries; the benefits should be great enough to justify the system's cost; and the system should accomplish the reduction in the most cost-effective manner available. Direct measurement of cost reductions would be difficult (although it would be possible and should be carried out), so evaluation of the improved system would also depend on measurements of secondary criteria correlated with the system's ability to reduce costs, primarily reliability and timeliness. Reliability is the capacity of the system to provide the required information accurately and in time to be useful to responders. An example of a reliability measurement is the roughly 25 percent failure rate of the present system in consequential highway accidents found in the incident case studies. To reduce the costs of incidents, any new system would have to improve the reliability of information enough to change the actions of emergency responders. The incident case studies indicate that emer- gency responders tend to take a cautious course of action, in part because they believe from experience that materials information can be unreliable. Responder actions may include precautions that have signif- icant costs (e.g., traffic blockages or workplace evacuations) but are believed to be warranted to reduce the risk of injury or death. Because of the cautious practices of emergency responders, and because in many circumstances responders do not have a wide range of options for dealing with an incident, incremental improvement in information reliability or timeliness would have no effect on emergency responder behavior in many incidents. On the other hand, an improve- ment in reliability great enough to boost overall responder confidence could result in more efficient response not just to incidents in which information failures would otherwise occur, but also in other incidents where responders now take costly precautions because they do not trust available information. Because the existing system's performance has not been measured, it is not known how much improvement in reliability, or timeliness is needed to reduce incident costs. This knowledge can be obtained only by evaluating the existing system and testing possible improvements.
Options To Improve Information: Technological Components 103 DESIGN OPTIONS A range of options exists for using information technology to give responders access to information at hazardous materials transporta- tion incidents. The range of choices is broad in part because the problem has lacked definition: it is not established what capabilities would have the greatest payoff in reducing the costs of incidents. Because various system designs may be feasible, with various capa- bilities and costs, the evaluation will be confronted with questions about the appropriate function and scale of applications. This section describes options for the design or architecture of systems, design issues, administrative issues, and the costs of a hypothetical technology application. System Architecture The original central reporting system proposal is one design for a system to provide vehicle-specific information to emergency responders at incidents. At least three other designs are conceivable: Phased approach to the central reporting system: The central reporting system proposal could be introduced in phases. The starting point could be a voluntary program rather than a regulatory mandate, and only those carriers and shippers who were best equipped to elec- tronically submit records would be included. The system would be limited to a subset of hazardous materials that pose special safety or identification problems. Access via a clearinghouse to shipper or carrier data bases: Access would be provided through a clearinghouse to carrier or shipper data bases that identify shipment contents. The connection could be auto- mated, using standards for communications. However, manual connec- tions (e.g., putting a responder in direct touch with a shipper or carrier employee with access to the company's shipment data base) might be more cost-effective, especially initially. No central data base of ship- ment records would be created. It would probably be necessary for carriers to maintain most shipment data bases to which the clearing- house had access, because only the carrier has direct knowledge of all handlings of a shipment.
104 HAZARDOUS MATERIALS SHIPMENT INFORMATION The clearinghouse approach would avoid the cost of transmitting and storing large quantities of data that were never used and some of the security problems of the central reporting system. It might be a way to take advantage of the growth in electronic data interchange and industry shipment-tracking systems. On-board identification systems: Shippers would place electronic transponders or transmitters on vehicles or shipments. They could function in several ways: automatic electronic identification tags could transmit messages when queried that identified the materials on board (electronic placards), or they could contain identification codes allow- ing emergency responders to ascertain the contents by contacting the shipper or carrier or a clearinghouse. These automatic electronic identi- fication applications could be tied to the corridor truck monitoring systems that are under development and to existing railcar automatic electronic identification systems. Alternatively, on-board transmitters could periodically broadcast alerts of accidents or the position (ob- tained via use of navigation satellites), identity, and contents of a vehicle. Commercial and government systems with functions similar to all of these are in operation today. The advantage of automatic electronic identification over visual placards and shipping papers is that it is remotely readable, so re- sponders could obtain vehicle identification or product information at less risk. A transmitter triggered by heat or acceleration sensors could be made to function despite a crash or fireby ejecting it from the vehicle or by having it broadcast a final message to a recording receiver before it was destroyed. Remote sensing would also have value in enforcement. Each option would include the following components: Shipper and carrier internal information systems required to main- tain records about the contents of vehicles; A central service facility whose functions could include maintain- ing a central shipment information data base; facilitating communica- tions with shippers, carriers, and responders; or maintaining other information resources; The interface with emergency responders (communications de- vices, hot lines, or other requirements); and Regulatory, enforcement, training, and evaluation mechanisms.
Options To Improve Information: Technological Components 105 Specification of options would address not only the technical means of recording and communicating information but also data definitions, responsibilities for reporting shipments, definition of shipments re- quired to be reported, rules governing data access, the schedule of implementation, and arrangements for finance, management, evalua- tion, and enforcement. The foregoing options provide vehicle- or shipment-specific informa- tion. Other technology applications could provide access to informa- tion that is not vehicle- or shipment-specific. Examples are stand-alone computers and software that provide data bases of materials properties or decision tools for managing incidents, central facilities such as CHEMTREC accessible by data or voice communication that would provide advisory services and would themselves have rapid access to various data bases and to shipper and carrier experts, and more com- plex networks giving access to all the parties available to assist in an incident and other data bases. Provision of vehicle- and shipment- specific information (by requiring real-time reporting of each haz- ardous materials shipment) is the defining characteristic of the original central reporting system proposal and would be the source of most of the complexities and costs of implementing it. Systems providing only non-vehicle-specific information would be much simpler. Design Issues The act calling for this study includes specific questions about design issues that would have to be resolved before the central reporting system or most of the other options could be implemented. This study has not been able to answer all the questions, because doing so would require actually carrying out the full design process outlined at the beginning of this chapter: analyzing needs and specifying and testing alternatives. Answers would require knowledge about infor- mation needs and the relation between system performance and benefits that is not available. Answers would also require cost- effectiveness analyses that could only be conducted on the basis of field testing of systems, which was not possible with the resources available for this study. These design issues, as well as some issues that were not emphasized in the act, are discussed next to highlight their importance for the
106 HAZARDOUS MATERIALS SHIPMENT INFORMATION success of any effort to automate hazardous materials emergency re- sponse information In formation Content The act asks what standards should be set for the form and content of records of shipments in an information system. As noted at the begin- fling of this chapter, determination of information needs is the primary design issue. As shown in Chapter 3, the data base of incident response experience required to determine needs does not exist. Not only must needs be specified in general, but also the exact format of information (for example, the coding schemes or names to be used to identify materials) must be determined. Answers to such questions would be dictated in great part by human factors considerations and could be discovered only by experimentation. Data Security Confidentiality is a major concern of shippers and carriers. Records of customers, sales, and shipments are regarded as proprietary be- cause of their competitive significance. A central reporting system could become a target for industrial espionage and other unautho- rized uses. Rules would be needed to protect data and to govern access to information through the system, especially access by private parties. System use would have to be monitored to detect unusual patterns of inquiries. Effectiveness of confidentiality safeguards would have to be tested. Coverage A system with broad coverage would contain information on at least every shipment that now requires placarding or carriage of shipping papers. As shown in Chapter 3, the severity of responder information problems differs greatly among categories of shipment and transporta- tion mode. The inherent danger posed by incidents varies according to the material and quantity. Costs of compliance with new information requirements would differ greatly among segments of the shipper and
Options To Improve Information: Technological Components 107 carrier industries. Therefore, evaluation must measure the potential for reducing cost while affecting benefits only slightly by targeting applica- tion to a subset of shipments. Human Factors The reliability of the data in an automated information system will depend on the design of the interface between the system and the people who provide the data. Input to a central reporting system could come from a large and diverse population of providers: employees of hun- dreds of thousands of companies ranging from the largest corporations to small businesses with little experience with information technology. The usefulness of the information would depend on the design of the interface for access by emergency responders. The incident case studies described in Chapter 3 showed that an important cause of failures in the present system is the difficulty responders face in interpreting information in unstandardized formats from multiple sources. To deal properly with human factors in an automated information system would require extensive testing of alternatives. Compatibility with Shipper and Carrier Internal Information Systems Considerations of reliability, cost, and coverage indicate that a practi- cal hazardous materials emergency response information system would have to be compatible with existing shipper and carrier practices and build on their existing and emerging information systems. A system in daily use that is essential to a company's operation will have much greater reliability than one maintained only to meet regulatory require- ments and rarely exercised. Avoiding duplicative requirements would also reduce costs. Building on existing capabilities is critical because of the rapid intro- duction of innovative information technology in transportation. Some carriers and shippers could comply today with many of the record- keeping and reporting requirements necessary to participate in a cen- tral reporting system or in the clearinghouse option at a cost that would not be deeply disruptive. Throughout transportation, carriers and shippers are developing capabilities for tracking shipments in transpor- tation to reduce inventory costs, to allow more efficient fleet utilization
108 HAZARDOUS MATERIALS SHIPMENT INFORMATION by monitoring vehicles, and to eliminate paper transactions by use of electronic data interchange. Computer-based information systems in use today in transportation that perform functions similar to some that would be required of a hazardous materials information system are described in Appendix D. To investigate how new hazardous materials information reporting requirements would integrate with existing information systems and procedures in some actual situations, interviews were conducted with selected shippers, carriers, and emergency response agencies. They are presented as case studies and summarized in Appendix E. The case studies showed that the relevant information system capabilities differ greatly by size of firm, industry, and mode. Rail is the most advanced, because automated car-tracking systems are widely implemented. Few firms in other sectors have information systems in place that routinely record the location of hazardous shipments or the contents of vehicles carrying hazardous materials in real time. Automatic recording of pickups and deliveries of shipments from trucks in real time would be critical to the function of a central reporting system, but it is only beginning to be implemented. Imposition of a central reporting system requirement would force nearly all firms to develop new information systems. However, firms with some shipment-tracking capabilities would face much lower costs in complying with requirements for an automated system. Matching Solutions to the Most Important Information Failures To improve performance, a system using information technology would have to solve the problems that lead to failures of the current system. Six categories of recurring failures of the existing system were identified in Chapter 3. Some ways in which technology applications might help avoid these failures are given in Table 4-1. The effectiveness of any of these possible solutions could only be determined through testing. This comparison of technology applications to actual problems sug- gests that technology probably is not the key to solving any of the problems, although it might play a complementary role. All of the technology applications given in the table would require complemen- tary provisions for regulation and training. Such new provisions alone, without the technology application, might substantially reduce the
Options To Improve Information: Technological Components 109 problem. For example, shippers would make more complete informa- tion available in a data base (see the third problem category in Table 4-1) only if regulations specified the information and format required. Once the additional requirements were established, the improved infor- mation would often be readily accessible to responders if it were simply printed on the shipping papers. Several of the suggested solutions could be implemented without the creation of a central reporting system or pointer system for access to vehicle-specific records. For example, emergency responder decision aids and automatic error checking in completing shipping documents could operate as stand-alone systems. Finally, none of the technology applications suggested would complement one of the most critical information needsâthe immediate, attention-seizing warning of a haz- ard that the placard supplies. Administrative and Evaluation Issues The legislative charge for this study includes a list of questions about administrative issues in implementing the central reporting system or alternatives. Administrative arrangements, including the method of funding the central facility, funding of local emergency response agency equipment and training needed to use an improved information system, choice of private versus contractor or direct public operation, questions of liability of the operator of the central facility or others for failures of the system, enforcement of participation, and the choice of voluntary or legally mandated participation would affect performance. Options for these arrangements could be compared in evaluation of start-up systems. Continuation of data collection after the system was operational to evaluate the improved system would be an essential component of administration. Cost of Technology Applications The cost of a new hazardous materials shipment information system would depend on the functions it supported, the extent to which it took advantage of existing systems, and the categories of shippers, carriers, and shipments covered. Cost estimates must be based on arbitrary
TABLE 4-1 Possible Information Technology Solutions to Information Failures Problem Category Specific Problems Possible Means of Alleviation - Required sources of information are missing or inaccurate Placards, shipping papers, or other information sources are obscured, destroyed, or inaccessible because of a crash or fire or threat of fire, explosion, or toxic exposure Information sources are in compliance with regulations and accessible, but they fail to efficiently convey important information to responders Placards missing or incorrect Shipping papers missing, incorrect, or incomplete Placards destroyed or obscured Shipping papers destroyed or inaccessible Dangerous placard not descriptive of hazards Multiple shipping papers that are cumbersome to review Responders did not trust all information sources available and therefore did not use them Automatic error checking on data entry Enforcement audits of shipment data bases Access to remote data base (when visual identification of vehicle is possible) Survivable transponder Receiver records final distress message of on-board transmitter Network gives responders access to more complete data (e.g., MSDSs, consists, expert response guidance) System organizes information to highlight key facts (e.g., summary manifest of mixed load, interaction hazards)
Information is insufficient because the material or shipment is exempt from some federal hazardous materials transportation regulations Some potentially harmful materials or shipments are not covered by DOT regulations governing placarding, labeling, and shipping papers because they are not designated as hazardous, do not meet minimum weight/quantity thresholds, or are not involved in a clearly transport- related activity Responders sometimes cannot readily confirm that a material is not hazardous System allows access to information on nonhazardous as well as hazardous shipments Driver uses the emergency response information system to become familiar with cargo Vehicle operator is unprepared to Truck driver unable to identify basic provide information hazards of cargo Train crew did not take proper actions to assist responders, such as identifying themselves and providing on-board documentation Responders fail to obtain or use Responders unaware of all Decision aids in system designed to available information information sources available or support inexperienced responders unprepared to take advantage of all available information
112 HAZARDOUS MATERIALS SHIPMENT INFORMATION assumptions about these factors, since the appropriate scale and func- tioning of the system could only be determined in an evolutionary process starting with test implementations. However, estimates can illustrate the likely magnitude of costs for a system of specified extent and show the sensitivity of costs to scale and the degree of reliance on existing information systems. The cost estimates below are derived in part from the shipper and carrier case studies and interviews with emergency response organiza- tions and information service providers described in Appendix E. Inter- viewees were asked about the costs of current activities that are in some way analogous to functions that would be required under a central reporting system. They were also asked to describe the new activities they would have to undertake to participate in a central reporting system and to estimate the costs of these activities. The estimates therefore are speculative and rely heavily on judgment. Illustrative cost estimates were carried out for a limited version of the central reporting system, with no central shipment data base but with a central facility that could link emergency responders with data bases maintained by carriers and holding records of vehicle contents. Carriers with large hazardous materials volumes would rely on automated data entry (e.g., bar-coded shipping papers) and data entry and transmis- sion from trucks at pickup and delivery points. In the base case esti- mate, a limited subset of hazardous materials shipments would be covered. Participants would include all railroads, 10,000 truck fleets, 200,000 trucks (an estimate of the number of tractor-semitrailers and heavy single-unit trucks that carry hazardous materials with some regularity), 100,000 shippers, and 30,000 local government emergency response agencies. The 100,000 shippers would be responsible for providing shipment information in prescribed formats. The estimate of the number of trucks excludes a much larger number of small and medium trucks that regularly carry hazardous materials and large trucks that rarely carry hazardous materials. Of the 200,000 trucks, 60,000 that carry hazardous materials most frequently would be equipped with two-way data communications and bar code equip- ment. About 40 million annual carload, truckload, less-than- truckload, and containerized hazardous materials shipments would be covered; this excludes a larger number of hazardous materials ship- ments carried primarily by small and medium trucks. This is a much more modest system than that envisioned in the original proposal since it excludes many shippers, carriers, and emer-
Options To Improve Information: Technological Components 113 gency responders. The example considers only the rail and truck modes, but a system might be extended to cover waterborne and air freight. The four major cost components would be shipper and carrier inter- nal information systems, the emergency responder interface, the central facility, and government enforcement and evaluation. Shipper Costs In the illustrative system, direct shipper costs would be relatively small because it is assumed that carriers would maintain the shipment data bases. (Of course, most carrier costs would be passed on to shippers.) Shippers might bear some costs to facilitate operation of the hazardous materials information system. For example, they might provide bar code labels on packages or increase the use of electronic data inter- change for transmission of shipping documents to carriers to facilitate data entry. If most shippers made no change in current practices but a small fraction incurred costs to develop or expand electronic data interchange capabilities, total direct start-up costs to the affected ship- per group might be in the tens of millions of dollars. Annual operating costs might be roughly the same. The shippers' costs might be partially offset by efficiency gained in day-to-day operations. Rail Carrier Costs Since nearly all significant railroads have some automated car-tracking capability, costs to railroads to participate in the illustrative system would also be limited. Each railroad would be required to develop software and communications systems to provide new information services to emergency responders via the central facility. Start-up costs might be a few hundred thousand dollars per railroad, and annual operating costs might be of similar magnitude. Industrywide, start-up and annual operating costs might each be $10 million. Rail costs would be limited because the data to support the system already exist. The railroads have probably spent hundreds of millions of dollars to develop this capability over the past two decades, and annual operating costs are probably at least in the tens of millions of dollars. A hazardous materials emergency response information system
114 HAZARDOUS MATERIALS SHIPMENT INFORMATION might be able to take advantage of this large investment for a small incremental cost. Trucking Costs Each truck carrier would have to develop an automated shipment- tracking system to comply with the requirements of the hypothetical system. Each movement of a shipment into or out of a truck at a shipper or receiver facility or trucking company dock would be recorded in a central data base as it occurred. Costs would include central computer facilities, data entry devices on docks and on board trucks, and commu- nications links. Operating costs would include maintenance of the system and any net increase in labor for data entry by clerks, dock workers, or drivers. A few large carriers, particularly in the less-than- truckload segment of for-hire trucking, have some of these capabilities. Remote data entry of waybill information by drivers at the time of pickup and delivery is just beginning to be used. The cost of developing a shipment-tracking system in a large less- than-truckload carrier, excluding costs of remote data entry by drivers at pickup or delivery, might be $10 million to $20 million. Numerous systems with varying capabilities for providing data communications between a central point and trucks on the road are available. Costs of such systems are from $500 to $4,000 per truck installation; annual communications and maintenance costs are $500 to $2,000 per truck. The total cost of the internal capabilities required in trucking to participate in a clearinghouse-type hazardous materials information system would be sensitive to the extent of coverage of the system. In the illustrative case (200,000 total tractor-semitrailers participating, of which 60,000 are frequent hazardous materials carriers), the cost for fleet headquarters and terminal facilities and in-truck data communi- cations for the frequent carriers could be about $2,000 per truck for start-up and $1,000 per truck for annual operating costs. Infrequent carriers might find it more economical to use systems that were less automated and had lower start-up costs. For example, a carrier that handled hazardous materials shipments infrequently could have drivers use cellular or public phones to call the dispatcher to report pickup or delivery of a hazardous materials shipment. At headquarters, the transaction could be entered manually in a shipment data base. Such a procedure might entail a start-up cost of a few hundred dollars per truck and a similar amount annually.
Options To Improve Information: Technological Components 115 Thus total costs for the illustrative system in trucking might be roughly $100 million to $200 million for start-up and a similar amount in annual operating costs. If participation were required of all trucks that ever carry hazardous materials, in particular of the numerous fleets of small and medium trucks that carry hazardous cargoes in local travel, total costs could be several times this amount. Costs for a trucking firm obviously would be sensitive to the extent of the company's existing shipment-tracking capability. For example, a less-than-truckload firm that already tracked shipments as they were handled in its terminals might have incremental costs for compliance that were half those of a carrier without automated shipment tracking. The few firms that are already beginning to record pickups and de- liveries of shipments by data transmissions from vehicles might save three-fourths or more of the costs of a similar company without such capabilities. Emergency Responder Costs The largest costs for emergency responders would be initial and on- going training. Responders might also need some communications and computer equipment for access to the central facility. Start-up labor and equipment costs might be $1,000 to $10,000 per agency, and the total cost might be $50 million to $100 million. Central Facility Costs The costs of some existing information services give a rough indication of the magnitude of costs for the central facility. CHEMTREC, a facility that offers emergency response information services very differ- ent from those the illustrative clearinghouse system would offer, has an annual budget of $3 million. AAMVANET, a service of the American Association of Motor Vehicle Administrators that provides rapid, on- line checking of state driver's license records to preveht truck drivers from receiving multiple licenses, has some similarities to the clearing- house system. It is a distributed data base, giving users transparent access to 50 state data bases. AAMVANET's annual operating cost is $7 million, and start-up costs were $15 million.
116 HAZARDOUS MATERIALS SHIPMENT INFORMATION Thus the central facility for the illustrative hazardous materials system might have start-up costs of $10 million to $20 million and annual operating costs of $10 million to $20 million. Summor' of Costs In addition to the preceding costs, the government would incur annual enforcement and evaluation costs of several million dollars. In total, if the illustrative system were adopted today, initial capital and start-up costs would be about $300 million, and annual operating costs would be about $250 million. This sum should be regarded as a rough estimate, since it is not known whether the scale of the hypothetical system would be appropriate or what the cost implications of attaining the reliability necessary to improve on the existing system would be. Even the unit cost estimates are speculative. However, the cost would be greater than the estimate of the economic costs of information failures in Chapter 3. Most of the costs of the hypothetical system would be for shipper and carrier information systems. Most carriers with shipment-tracking ca- pability and other relevant communications and information systems in place could avoid most of the costs that would be incurred by comparable carriers with no such capabilities. Since these capabilities are expanding, the net cost of implementing the hypothetical system would decline over time. However, for the foreseeable future, nearly all shippers and carriers would incur the costs of developing new procedures to keep continuous records of vehicles in which shipments are contained and to expand electronic data inter- change capabilities. Within a sector, the requirement of participation would disadvan- tage small firms. Development of the internal systems needed would be in part a fixed cost and, therefore, a smaller percentage cost increase for large firms. EVALUATION WITH PROTOTYPES Undoubtedly, every carrier that today has a shipment-tracking system or that uses in-vehicle communication or positioning systems in its trucks began to implement the system with a prototype installation or
Options To Improve Information: Technological Components 117 limited start-up. Evaluation of the initial installation led to refinements in design and was the basis for the final decision for full-scale imple- mentation. In developing a nationwide hazardous materials informa- tion system, which would be more complex than any one company's system, the same process would be essential. Prototypes of hazardous materials information systems would be controlled experiments with independent and rigorously designed eval- uation protocols. Their design, conduct, and evaluation should be governed by cooperative arrangements between governments and par- ticipating shippers and carriers. Models for such joint efforts may be found in several of the intelligent vehicle-highway system prototype implementations under way in the United States. If a prototype is demonstrated to have benefits sufficient to justify its costs, it should become the foundation of a widely implemented system. The resulting systems would not all necessarily be legally mandated; some could be voluntary industry-supported activities. Examples of possible proto- type implementations are Emergency responder access to shipment-tracking systems of car- riers via an established network service, Incorporation of hazardous materials monitoring functions in truck automatic electronic identification corridor demonstration proj- ects, and Emergency responder access to the rail industry car-tracking and waybill data system. These three examples are outlined in this section, and a limited start-up of a rail industry system is considered in more detail and proposed as a practical initial step. Network Service-Based System Several private firms offer electronic data interchange network services to shippers. In a typical application, a shipper wishes to obtain periodic status reports on all its shipments in transportation. The shipper in- structs all the carriers handling its freight to transmit periodic status reports to a third-party service. The service consolidates the reports, relays them to the shipper, and performs other functions, such as preparation of customized management information reports.
118 HAZARDOUS MATERIALS SHIPMENT INFORMATION The prototype would be a system that allowed emergency responders to contact the third-party service and obtain waybill information for a specified trailer, railcar, or container. Carriers would periodically trans- mit records for each hazardous shipment in transportation to the network service, which would maintain a central data base. Alter- natively, the network service would query carrier data bases (either automatically or with a human intermediary) in response to an emer- gency responder request. The advantage of the approach is that at least part of the information required by emergency responders is already maintained by carriers with shipment-tracking systems and handled in electronic data interchange transactions via these services, so standard formats and protocols and an infrastructure exist. Although many of the basic capabilities for meeting emergency responder needs exist, new features would have to be developed for the prototype. In particular, the method selected for ensuring access to current shipment records keyed to vehicle number would require a substantial investment in systems development. For example, carrier transmittal of a waybill on request or shipment-tracking queries keyed to vehicle number may not be available on current network services, and no existing service maintains a central data base of shipment information. In the prototype, trailers would be identified by the exist- ing trailer identification numbers used by motor carriers. These numbers are now used to identify trailers in the carriers' shipment- tracking systems. Emergency responder contact with the prototype would be via per- sonal computer or voice telephone. New standard messages and a software interface would have to be developed for use by responders. Evaluation should include comparison of the utility and cost of a fully automated link of responder and shipment data with a link through a human intermediaryâa shipper, carrier, or network service employee who had access to shipment data bases. The objectives of the prototype would be (a) to determine the extent to which emergency responders can rely on existing facilities for elec- tronic transmission of shipping information, (b) to demonstrate feasibility, (c) to obtain data on operating costs, and (d) to observe the value of the information in emergency response situations. Evaluation of the performance of the system in emergencies would depend primar- ily on drills, since the number of actual incidents that could be expected to occur among the participants during a reasonable pilot period might be very small.
Options To Improve Information: Technological Components 119 Automatic Electronic Identification Corridor Demonstration At least two projects are under way demonstrating applications of automatic electronic identification technology for reducing the enforce- ment and compliance costs of collecting taxes on trucks and regulating truck size and weight, safety, and other aspects of the industry. The Crescent Demonstration of the Heavy Vehicle Electronic License Plate Project involves primarily western state transportation departments and truck operators, and the newer Advantage 1-75 project is a similar program in the eastern United States. The demonstrations have three components: fleets of trucks volun- tarily equipped by their private-sector operators with automatic elec- tronic identification transponders, a major truck corridor equipped with transponder readers and communications lines at weigh stations and other key points, and a central facility linked to the reader station that maintains a data base on trucks in the corridor. The applications planned are to reduce costs by avoiding multiple stops for safety inspections, weighings, and checking of permits at border crossings. The goal of these projects is to link systems nationwide and integrate them with other automatic electronic identification applications, in- cluding shipment- and equipment-tracking functions and automatic toll collection, once feasibility and value are demonstrated. Addition of a hazardous materials shipment-tracking component could take several forms. In the simplest, hazardous cargo waybill information would be entered in the truck's record in the central data base when other information about permits and safety inspections was entered. Emergency responders would have access to the central data base by voice or data linkage and would idditify trucks by DOT number and trailer number, a special visible ID developed for the demonstration, or license plate number. This system would take advan- tage of the data base, but not the transponder. The prototype could also test the utility of transponders in emer- gency response by equipping emergency responders with portable readers. Responderwould use the truck identification read from the transponder to locate the truck in the data base. Alternatively, a trans- ponder of the read-write type could be encoded with information about the hazardous cargo in the vehicle. The major difficulty that the prototype would confront would be keeping records up to date. A truck entering the corridor may travel some distance before reaching the first station where it could enter
120 HAZARDOUS MATERIALS SHIPMENT LNFORMA11ON required data in the data base. Similarly, if a truck reentered the corridor after a pickup or delivery, its record could not be updated until it reached the next station. Simple techniques for entering cargo information in the central data base would have to be developed, because keyboard entry by an inspection official or driver would be time-consuming and subject to error. Possibilities include electronic data interchange between the carrier and the central data base at the time of dispatch of the truck or use of bar code or "smart card" electronic label technology. If trucks were equipped with read-write transponders encoded with cargo infor- mation, they would be read to enter cargo information in the central data base. The prototype would be an added component of an established automatic electronic identification corridor demonstration project. These projects involve the joint participation of state departments of transportation along the route of the corridor, federal agencies, and motor carriers. The projects have contractor support for systems devel- opment and operation. Participation of emergency response agencies along the corridor would also be required. A component of this prototype could be demonstration of trans- ponders capable of surviving crashes or fires. Options include a small radio transmitter that would be ejected from the vehicle by a mecha- nism actuated by heat or shock and that would broadcast an identifying message on a frequency known to responders; a hardened transmitter that could survive shock and heat for some period; or a system involv- ing a transmitter that would broadcast a final message triggered by sensors, together with a network of recording receivers. Evaluation of the technical feasibility and cost of these devices would be based on the prototypes. Estimation of benefits and determination of whether the costs would justify the use of these systems on any category of vehicles would be based on analysis of past incidents. Emergency Responder Access to Rail Industry Car-Tracking Data Base A limited start-up of a rail industry system would be a practical first step. It would be straightforward to organize because only a small number of carriers would be involved and because most of the system components already exist. It would demonstrate a service that emer-
Options To Improve Information: Technological Components 121 gency responders would be likely to find valuable and provide an experimental setting for addressing many of the organizational, techni- cal, and cost-effectiveness issues that any system would face. Concept The rail industry maintains an information system, TRAIN II, that continuously keeps account of the location and status of railcars, trailers and containers moving on rail, and waybills for the shipments moving in those cars for most rail traffic in the United States. The system supports electronic communication to update or query the data base by operators, car owners, and other authorized users. The system could enable emergency responders to obtain immediate information electronically on the contents of nearly any railcar they encountered in a wreck or hazardous materials incident, if a means were provided to link them to the data base. Such a linkage could be provided directly by the operator of the rail information system or by a third party. In the first phase of a limited start-up, it would be most efficient to involve only one railroad and give participating responders in the railroad's territory access to its car-tracking data base. If evaluation of the first phase showed that worthwhile performance improvements had been achieved, the second phase would expand the test to a nationwide trial giving responders access to the industrywide data base jointly maintained by the major railroads. The limited start-up could also compare alternative communications links between responders and the rail data base and alternative information resources for interpreting the cargo information received. Design Considerations Responder Interface. In the simplest system, the responder would dial an 800 telephone number, enter the identification number of a railcar on the telephone push buttons, and hear a synthesized voice speaking the Standard Transportation Commodity Codes (STCCs) of all commodi- ties on the car. (STCCs are standard codes that provide fairly detailed information about the commodity.) This might be adequate for a single car carrying one commodity but would be cumbersome for retrieving the contents of an intermodal container or of several cars in a train.
122 HAZARDOUS MATERIALS SHIPMENT INFORMA11ON Therefore, the limited start-up could also test a data link that would transmit a list of commodity codes to a responder's computer or a capability to transmit a commodity list by fax. The limited start-up could address means of efficiently giving responders access to commod- ity information for an entire train. Tests should also compare an automated system with an arrange- ment giving fire fighters direct contact with an expert railroad em- ployee who has access to railroad data. Information Provided. The rail data bases identify commodities by STCC code. It might be possible to modify the data bases to include the waybill proper shipping name. The consist, a list of the cars making up an entire train that is sometimes useful to emergency responders, is not available through the industry joint car-tracking data base. Interpretation of Commodity In formation. The limited start-up could test alternative means of interpreting the STCC codes and deciding on proper means of dealing with an incident. The options include use of a printed STCC reference text (such a text is used by some emergency response agencies) together with other printed materials such as the DOT Emergency Response Guidebook, CHEMTREC, or a stand- alone computer-based decision aid. It would also be possible for the railroad or third party to interpret the STCCs and provide a longer text message. Coverage. Gaps in coverage would include cars on some short line railroads, especially piggyback trailers, and possibly cars that had been loaded or unloaded within the previous several hours. Problems may be encountered with the completeness of records of the contents of inter- modal containers and trailers. Security. The system would have to incorporate safeguards against loss of proprietary data on shipments. Evaluation Rail hazardous materials incidents are infrequent, so evaluation would require drills. Evaluation would include measurements of
Options To Improve Information: Technological Components 123 Reliability of cargo identification compared with existing means for bulk and intermodal cargoes and for single- and multiple-car incidents, Speed of identification, Effectiveness of alternative responder interfaces, Effectiveness of alternative means of interpreting commodity in- formation, Responder reactions (willingness to use the system, training re- quirements, and effect on responder decisions), Effectiveness of security, and Costs. In the simplest test, responders would observe a car on the rail line and use the system to identify its contents. When the car reached its destina- tion, the contents listed on the waybill and the actual contents would be verified. More complex tests would involve simulated incidents. In a simula- tion the ability of rail personnel and responders to communicate through standard means (e.g., shipping documents and placards) would be compared with their success using the system. Evaluation should consider the potential for benefits of the system beyond its primary objective of improved identification of materials contained in vehicles involved in incidents. Such benefits might include other applications of data communication network access for emer- gency responders, identification of materials regularly transported through an area for emergency preparedness, and applications in en- forcement of hazardous materials and hazardous waste regulations. Uses for enforcement or for monitoring materials flow will be contro- versial among shippers and carriers. Evaluation will have to consider the practicality as well as the value of such applications in a system whose success will depend to some extent on shipper and carrier. willingness to participate. During the course of a 2-year limited start-up on a large railroad, perhaps 10 actual incidents involving public-sector emergency re- sponders would occur. Experience during actual episodes might pro- vide further evaluation information, provided all responders in the railroad's territory had been notified of the availability of the test service.
124 HAZARDOUS MATERIALS SHIPMENT INFORMATION Participants and Governance The participation of all parties in the limited start-up would be volun- tary. Participants would be the railroads, possibly car owners, and emergency responder agencies. The first phase could involve one rail- road and perhaps 20 to 50 responder agencies in the railroad's terri- tory. The federal government's role would be to recruit participation, fund local emergency responder costs and test costs, and observe the tests. Testing could be conducted by the government or by an indepen- dent organization established by the participants. Costs The major costs would be incurred in modifying existing rail industry information systems, equipping participating emergency responders with computer and communications hardware and software, training responders and railroad personnel in the use of the system, and con- ducting the tests. A 2-year limited start-up involving one railroad and 30 emergency response agencies would cost roughly $4 million. Ex- panding the test to a nationwide trial in the third year would raise the total cost to roughly $7 million (Table 4-2). TABLE 4-2 Cost Estimate for a Rail Shipment Information System Prototype Cost ($ millions) Phase 1 Phase 2 Total Responder equipment and training 1.0 1.0 2.0 Railroad system development 0.5 0.5 1.0 Management and systems integration 0.5 0.5 1.0 Testing 2.0 1.0 3.0 Total 4.0 3.0 7.0 Nom: Phase 1 is a 2-year test involving one railroad and approximately 30 responders. Phase 2 is a third year involving the joint industry car-tracking system and approximately 90 responders.
5 Options To Improve Information: Nontechnological Components The nontechnological means of improving the information avail- able to emergency responders at hazardous materials transportation mci- dents are identified in this chapter. Improvement can be achieved through changes in regulatory requirements, enforcement, and training. The technological options for improving information described in Chapter 4 and the nontechnological means of enhancing the existing system described in this chapter are related. To evaluate technology- based systems, it is necessary to consider whether improvements to current practice could achieve objectives more efficiently. Effective regulations, enforcement, and training will be the foundation of any improvement in emergency response, including improvements that take advantage of new technology. The most serious deficiencies in regulations, enforcement, and training would have to be addressed before technology applications could be expected to pay off, and any technology applications adopted would necessitate new, complemen- tary regulations and training and enforcement activities. This chapter is organized into three sections covering regulatory, en- forcement, and training options. Information problems identified in Chapter 3 from the case studies and responder interviews are cited as examples of problems that the options are intended to address. In the summary at the end of this chapter, the potential improvements are correlated with the six categories of information problems described in Chapter 3. 125
126 HAZARDOUS MATERIALS SHIPMENT INFORMATION The list includes improvements suggested by the incident analyses and proposals from other sources, including emergency responders interviewed and National Transportation Safety Board (NTSB) recom- mendations. It was not possible to assess the costs and benefits of each option. However, each merits consideration. If the options proposed are determined to be inadequate, the U.S. Department of Transporta- tion (DOT) should devise other means of correcting the information problems identified in this study. Many of the options described in this chapter, including those deal- ing with placards and shipping papers, are being addressed in DOT rule making (Docket HM-206, 57FR24532, June 9, 1992). Several of the options were considered in past rule makings but were not accepted or acted upon. The rule making process is required for DOT to enact regulations. However, rule making alone, unsupported by an integrated program of data collection, research, and analysis, is not adequate for evaluating some proposals. Furthermore, rule making is not well suited for generating input from local emergency respondersâa rule making that depended on information submitted to the docket by local agencies on their own initiative would be unlikely to obtain comprehensive and unbiased data about these agencies' circumstances. A more direct and objective method of obtaining input from emergency responders is required to evaluate the options proposed. DOT needs a systematic procedure for identifying problems with existing practices, identifying solutions, and evaluating and implement- ing the solutions. The procedure must start with monitoring of the performance of the existing system, as described in Chapter 3 and Appendix B. Generation of options can rely on rule making as well as active outreach to emergency responders. In this context, evaluation means estimates of costs and benefits of new regulations and enforce- ment and training efforts on the basis of tests and analyses of historical incident data. REGULATORY OPTIONS Placarding Dangerous Placard DOT placarding regulations refer to two broad categories of hazardous materials, known as Table 1 and Table 2 materials (49 CFR 172.500).
Options To Improve Information: Nontechnological Components 127 Table 1 materials are considered by DOT to present the greatest danger in transportation. The regulations require any quantity of these mate- rials to be placarded, using the specific placards from the hazard class or division assigned to the material. Table 2 covers materials that are considered by DOT to present a lesser danger in transportation. In general, shipments of Table 2 materials that have an aggregate weight of 1,000 lb or less are not required to be placarded. If a vehicle is carrying more than one shipment of Table 2 materials and none weighs more than 5,000 lb, a Dangerous placard may be displayed in place of specific placards assigned to the material's hazard class or division. Interviews with responders and findings from the case studies indi- cate that the Dangerous placard does not adequately convey hazard information. For instance, thousands of pounds of materials may be contained in a vehicle placarded Dangerous, yet responders could not determine from the placard the general types of hazard present (e.g., flammable or corrosive). Placards for specific materials or groups of materials provide this hazard information through hazard symbols, coded colors, material identification numbers, or some combination, but the Dangerous placard does not. Elimination of the Dangerous placard would require placarding of each hazard class or division in a shipment when the aggregate weight exceeded a specified threshold. Whereas this might address some emer- gency responder concerns, it might create confusion if vehicles were required to display large numbers of placards. The change also would entail costs to industry. Shippers and carriers would have to purchase and maintain larger supplies of specific placards and provide additional placard holders on vehicles. Drivers would have to observe more closely the types and quantities of hazardous materials on the vehicle and change placards as necessary when materials were loaded and unloaded. The 1990 Hazardous Materials Transportation Uniform Safety Act (HMTUSA) requires DOT to evaluate methods for improving the current placarding system. The case studies and responder interviews indicate that the Dangerous placard is a major concern of responders and a potential source of confusion in the event of an incident. To address these concerns properly, a formal study of the effectiveness of the Dangerous placard and its alternatives is required. Such a study requires input from emergency responders and recognition of the im- portance of placards in providing hazard type information.
128 HAZARDOUS MATERIALS SHIPMENT INFORMATION Placard Visibility and Survivability The concern that placards may be obscured or destroyed during an accident, especially in a fire, is often expressed by responders. HMTUSA directs DOT to study methods of improving placard surviv- ability. However, the case studies did not indicate that failures in placard visibility and survivability are common problems. In cases where fire is involved, placards are sometimes the only source of hazard information available to responders, because shipping papers may not be accessible and the driver may have been seriously injured or killed. Prolonging the integrity of placards, even for a short time after exposure to flames, may be vital in some cases. There may also be techniques to make placards more visually striking and more visible at night and in other adverse conditions. In its evaluation of placard construction materials, DOT should balance cost considera- tions against the likely benefit of improvements in placard visibility and survivability. Placarding Quantity Thresholds Another option for improving placarding is reconsideration of the minimum quantity threshold for Table 2 materials. As discussed previ- ously, placarding is not required for Table 2 materials when the aggre- gate weight is 1,000 lb or less. This threshold is arbitrary in the sense that it is imposed on a diverse group of materials that have different hazards. Emergency responders on the committee expressed concern that certain Table 2 materials, when transported in any quantity, may be sufficiently hazardous to warrant placarding and reclassification to Table 1. Materials that were of most concern are given in the following table: DivisionPlacard Name 1.4 Explosives 1.4 1.5 Explosives 1.5 4.1 Flammable solid 4.2 Spontaneously combustible 5.2 Organic peroxide 6.1 (PG I or II) Poison (See the description of hazard classes and divisions in Chapter 2.)
Options To Improve Information: Nontechnological Components 129 Setting a lower threshold for placarding could affect hundreds of thousands of small shipments, some of which pose minimal danger. An extremely low threshold could result in the placarding of so many vehicles that the significance of placards is diminished. Current regula- tions lack a clear rationale for the quantity threshold. DOT should consider alternatives to the threshold weight for placards and should develop a rationale that explicitly recognizes the likely effect of the alternatives on emergency response operations. Placards and Labels on Vehicles and Containers at Fixed Sites Hazardous materials packages and containers in transportation are required to have markings and labels warning of hazards, and trans- port vehicles are required to be placarded. When transportation is completed, different labeling and marking requirements are imposed by the Occupational Safety and Health Administration of the Department of Labor (OSHA), which regulates hazardous materials in the workplace. In the Hazardous Materials Transportation Act of 1975, "transportation" is defined to include storage that is "incident to transportation"; however, the term "incident to transportation" is subject to interpretation. In practice, the determination of which requirements apply is subjective, especially under the following conditions: Hazardous materials are loaded into packages but are not labeled, and containers and vehicles are not placarded, until transportation is ready to take place. The loading process may occur over several days. Trailers may be used for temporary storage of hazardous materials without being placarded. Packages containing the hazardous materials within the trailers also may not be marked or labeled according to DOT requirements. Hazard warning labels may be removed from packaging delivered at the destination even though the package still contains hazardous materials. Hazardous materials may remain in a trailer for storage, after delivery at the destination, at which time placards may be removed even though hazardous materials are present. In many respects these gray areas are results of the jurisdictional differences of the two agencies; yet the regulations are intended to serve one group of users, emergency responders. Emergency responders pre-
130 HAZARDOUS MATERIALS SHIPMENT INFORMATION fer a consistent set of information sources in both transportation and fixed-site situations. The greater the number of alternative marking, labeling, and placarding arrangements, the more difficult it is for responders to recognize and interpret the various information sources, especially in an emergency. HMTUSA addressed this problem by directing the Secretary of Labor, in consultation with the Secretaries of Transportation and Trea- sury, to issue standards requiring any employer who receives a package, container, motor vehicle, rail freight car, aircraft, or vessel which contains a hazardous material and which is required to be marked, placarded, or labeled in accordance with regulations issued under the Hazardous Materials Transportation Act to retain the markings, placards, and labels, and any other information as may be required by such regulations on the package, container, motor vehicle, rail freight car, aircraft, or vessel, until the hazardous materials have been removed therefrom. (HMTUSA, Section 29) The provision covers only hazardous materials that are "received" and, therefore, does not consider materials being loaded or in storage before shipment or after delivery. Other actions that might alleviate the regulatory gray areas would be requiring that (a) hazardous materials be marked and labeled according to DOT requirements at the time of packaging for transportation; (b) trans- port vehicles used for storage be placarded according to DOT require- ments until all materials are removed from the vehicle; and (c) hazardous materials be placarded and labeled according to DOT requirements if a transport-type vehicle (e.g., semitrailer) is used for storage of the material, whether or not the vehicle is used for transportation. Consideration of these proposals requires extensive DOT and De- partment of Labor coordination. The objective is to ensure that emer- gency responders' needs are met in the most efficient manner. The importance of consistency in hazardous materials information should be recognized by both federal agencies. Shipping Papers and Other On-Board Documents Consist of Hazardous Materials in Motor Vehicles The case studies and interviews with emergency responders pointed out several problems related to multiple shipping papers. A common prob-
Options To Improve Information: Nontechnological Components 131 lem with trucks carrying mixed loads was that responders often could not quickly identify the hazardous materials because of the numerous papersâoften with varying formatsâthat had to be retrieved and reviewed. Current DOT regulations require that shipping papers for hazardous materials either appear first in the full set of papers (if nonhazardous materials are in the cargo) or be distinctively tabbed so that they can be quickly recognized by responders (49 CFR 177.817). The problems encountered by responders in the field suggest that these requirements may be insufficient or that compliance may be inadequate. One option to address this problem is a special summary consist used when two or more hazardous materials are carried on board a motor vehicle. Such a document would be prepared by the company operating the truck. The document would be placed on top of all hazardous materials shipping papers and would have the following characteristics: A format uniform among all motor carriers and containing infor- mation on the hazard class or division, packing group, and total weight of each hazardous material; Distinctive markings and a prominent heading; and Provision for additions and deletions of shipments as hazardous materials are picked up or delivered. Such a document would allow responders to quickly review the ship- ments before individual shipping papers are studied for more detailed information. Carrier 24-hr Emergency Telephone Number In some of the case studies in which a truck of a for-hire carrier contained shipments of several shippers, responders had to call several shippers to verify information in the shipping papers. Had the carrier been contacted, the calls may still have been required, but the carrier might have been able to assist in contacting the shipper and confirming shipping paper information. A 24-hr carrier telephone number would be most valuable when shipping papers are destroyed or otherwise inaccessible, because under these circumstances the carrier may be the only source of cargo and shipper information available to emergency responders.
132 HAZARDOUS MATERIALS SHIPMENT INFORMATION DOT requires shippers to list a 24-hr emergency contact on the shipping papers (49 CFR 172.604). Although there is no similar re- quirement for carriers of hazardous materials, HMTUSA requires DOT to consider extending this requirement to carriers by studying the feasibility, necessity, and safety benefits of requiring carriers involved in the hazardous materials transportation industry to establish contin- ually-monitored telephone systems equipped to provide emergency re- sponse information and assistance with respect to accidents and incidents involving hazardous materials. (HMTUSA, Section 26) Many shippers use CHEMTREC or another third party to fulfill the telephone monitoring requirement. NTSB has recommended that rail- roads maintain a 24-hr telephone number and point of contact for public and emergency service use in the event of an accident (NTSB Recommendation R-89-029). Larger railroads may be able to maintain their own 24-hr numbers; it is less likely that small truck operators could do so. Third-party services probably would be used by many motor carriers as the central contact point. As the shipper and carrier case studies described in Chapter 4 showed, most truck operators do not maintain central up-to-the- minute records of the contents of all their trucks. Responders who called for cargo information and response guidance would often be disappointed. During consideration of a carrier 24-hr telephone number, DOT should clearly define what information the system should be able to convey to responders and determine whether the information is adequate from the standpoint of responders and feasible to provide from the standpoint of carriers. The emergency responders interviewed reported that shipper 24-hr numbers sometimes fail in their purpose because the number is not answered or because the third-party answering service is unable to make timely contact with the shipper. Carrier 24-hr numbers would have the same potential problems and would require strong enforce- ment to be effective. Standard Transportation Commodity Code Numbers on Shipping Papers Railroads generally require that all movements of materials, both hazardous and nonhazardous, be described by material name on ship- ping papers, accompanied by a Standard Transportation Commodity
Options To Improve Information: Nontechnological Components 133 Code (STCC) number. This practice has also been adopted by some tank truck companies. From the standpoint of emergency response, an advantage of uni- form STCC numbers on shipping papers is that they provide another source of material identification. In practice, this advantage may be offset by the difficulty of ensuring the use of STCC numbers on interna- tional shipments, the cost of accessing STCC information, and the possibility that STCC numbers may be confused with other emergency numbers on shipping papers. The advantages and disadvantages of using STCC numbers on hazardous materials shipping papers for modes other than rail may need to be evaluated by DOT. Material Safety Data Sheets As discussed in Chapter 3, responders depend on shipping papers as a primary source of shipment information to identify materials in a vehicle. They then obtain material-specific details from other sources, such as the Material Safety Data Sheet (MSDS) required by OSHA. Suppliers must provide users with an MSDS for certain materials describing the hazards of the material and providing information on how the material can be safely handled, used, and stored (29 CFR 1910). In several of the case studies, the MSDS was faxed by the shipper or CHEMTREC to emergency responders and was a valuable source of information. Federal regulations require hazardous materials shipments to be accompanied by emergency information describing immediate hazards to health, risks of fire and explosion, preliminary methods for handling fires and leaks, and first aid procedures (49 CFR 172.602). This information may be provided in the shipping papers or in documents accompanying them. Shippers and carriers often use the DOT Emer- gency Response Guidebook, and sometimes the MSDS, to meet these requirements. The MSDS is familiar to many emergency responders, who use it at both fixed-site and transportation accidents. MSDSs are the principal means of hazardous materials information exchange between local facilities (other than transportation facilities) using or storing the mate- rials and the Local Emergency Planning Committees that have been created by local governments in accordance with the requirements of the Superfund Amendments Reauthorization Act of 1986. However,
134 HAZARDOUS MATERIALS SHIPMENT INFORMATION the format of the MSDS is not standardized, resulting in MSDS docu- ments varying in length from 1 to 20 pages and in content depending on the manufacturer of the material. MSDSs tend to emphasize informa- tion on storage and use hazards rather than transportation hazards. For instance, there is no requirement that the document contain informa- tion on a material's DOT hazard class or characteristics when packaged for transportation. Attempts are being made to standardize the MSDS to make it more useful under a wider variety of emergency situations. OSHA has initiated a rule making on the advantages and disadvantages of a standardized format from the standpoint of international as well as national harmonization (55FR2166, Jan. 22, 1990; S5FR20580, May 17, 1990). The Chemical Manufacturers Association has developed a voluntary preparation guideline for MSDSs submitted by shippers to CHEMTREC. Such a voluntary effort will require widespread cooper- ation to work. Hence, regulatory changes may be necessary to ensure compliance and uniformity. Any effort to standardize MSDSs should entail substantial emergency responder input. NTSB has recommended that DOT require the MSDS, or a similar product-specific emergency response document, to be attached to the shipping papers of bulk hazardous materials (NTSB Recommendation 1-83-002). DOT has not required use of the MSDS but has used it as an example of how shippers of hazardous materials may satisfy the re- quirement for emergency response information. As the process of stan- dardizing the MSDS proceeds, DOT should seek to ensure that the document addresses the information needs of emergency responders at transportation incidents. Other Issues Trailer Identiflcation Numbers Uniform numbering of truck trailers, similar to the identification sys- tem that is standard for railcars, piggyback trailers, and intermodal containers in the railroad industry, might help responders identify the owner of a vehicle and obtain cargo information. Uniform numbering was adopted to serve business purposes of the railroads and later proved beneficial to emergency responders during rail incidents. No similar business demand among motor carriers has
Options To Improve Information: Nontechnological Components 135 necessitated such a numbering system for trailers. If such a system were imposed on the trucking industry, the costsâdevelopment of the numbering scheme, initial marking, and maintenance of the numbersâpresumably would not be recovered by business uses as in the rail industry. To evaluate a proposal to require uniform trailer numbering, DOT would have to assess the contribution to improved emergency response that numbering could be expected to make and weigh that contribution against the costs. Information Overload To be useful, information sources need to convey hazard information to emergency responders in a clear and concise manner. The regulations that define hazardous materials and govern placarding, shipping pa- pers, and labeling need to be straightforward, so carriers and shippers can understand what is required and responders can understand and have confidence in the information sources. DOT should consider the impact on emergency responders whenever changes are made to regulations that may affect sources of emergency information. As an example, during the 1980s DOT incorporated in the definition of shipments subject to DOT regulation (49FR101, App.) a new list of hazardous substances defined according to the Environmen- tal Protection Agency (EPA) risk criteria that differ from DOT risk criteria for hazardous materials. When such changes are made, full consideration should be given to the potentially confusing effect on emergency responders from inclusion of substances with very different risk characteristics from those of the original list. A further example is the EPA-required manifest for hazardous waste, which is carried with shipping papers. Much of the information in the EPA manifest is intended for enforcement rather than emergency re- sponse, to track hazardous waste shipments and to ensure that trans- porters deliver the waste to certified disposal sites. The additional EPA-required information conceivably could diminish the ability of responders to identify materials that are immediate hazards in a trans- portation spill or crash and locate the relevant hazard information. The existence of differing state manifest requirements for hazardous wastes is a further complication. EPA has recently addressed the need for a uniform hazardous waste manifest (57FR24765).
136 HAZARDOUS MATERIALS SHIPMENT INFORMATION Complex and burdensome regulations may reduce compliance on the part of shippers and carriers (knowingly or unknowingly), which may result in missing or confusing information. A forum for fed- eral, state, and local government interaction could help avoid such problems. One of the potential problems with STCC numbers, DOT identifica- tion numbers, and other numbering systems is that they could detract from the standard United Nations/North American four-digit number- ing system currently used to identify hazardous materials in transpor- tation, thereby confusing emergency responders. Other examples of potentially confusing numbering systems include the EPA hazardous waste identifier under the Resource Conservation and Recovery Act (40 CFR 262-263) and the numbering scheme developed by the National Fire Protection Association to identify degrees of hazard, through indices of reactivity, flammability, and health hazard. Though DOT prohibits the placement of nonessential information, such as advertising, near required package markings and labels (49 CFR 172.304), it has not always fully considered the effect of changes in regulations and accompanying information on the ability of emer- gency responders to interpret hazards and efficiently locate vital infor- mation. Before regulations are changed, DOT must determine these effects and ensure that response capabilities are not hindered. ENFORCEMENT OPTIONS Evidence from the case studies and concerns expressed by emergency responders indicate that compliance with regulations concerning placarding, shipping papers, and other information sources may be less than adequate. There are several possible reasons for violations. Some may occur because carriers and shippers are unaware of or confused by the regulations. On the other hand, some may be delib- erate, because carriers and shippers are trying to avoid special rout- ing requirements (e.g., around tunnels and city centers) or additional transport charges (imposed by some carriers). To tailor the most effective enforcement program, the reasons for violations need to be well understood. The following are some enforcement-related actions that merit consideration.
Options To Improve Information: Nontechnological Components 137 Monitoring of Compliance There is no systematic nationwide monitoring of rates of compliance with hazardous materials transportation regulations. The principal existing data are the frequency of citations. Frequency is not a reliable indicator because enforcement practices vary widely by jurisdiction and because vehicles inspected are not a representative sample of the population. Compliance monitoring should be based on random in- spections conducted according to uniform procedures. Without such data it is difficult to gauge the extent of noncompliance, set priorities, or measure progress. Enforcement Priorities; Fines and Penalties HMTUSA established a minimum civil penalty ($250) for violations. It is not known whether the minimum penalty will deter intentional violations or ensure that adequate efforts are made by industry to understand and observe the regulations. A higher minimum penalty might promote diligence on the part of shippers and carriers in prepar- ing and ensuring the accessibility of shipping papers and placards. To be most effective, minimum fines and enforcement priorities should be consistent from state to state. Specific kinds of violations that should be considered for priority enforcement, including higher minimum fines and penalties, are dis- cussed in the following subsections. Missing or Incorrect Placards State inspection data cited in Chapter 3 suggest that placard regula- tions are violated in roughly 30 percent of hazardous materials shipments by truck. Missing or incorrect placards did not account for a disproportionate share of the problems in the case studies, but the problem is a serious concern of emergency responders. Placards provide the first warning of hazardous materials, and missing or inaccurate placards are a far greater immediate danger to responders than most other information deficiencies. In this respect, they may merit high minimum fines.
138 HAZARDOUS MATERIALS SHIPMENT INFORMATION Accessibility and Format of Shipping Papers Incidents in the case studies involving trucks carrying mixed loads of hazardous materials with multiple shipping papers commonly had information problems. The large number of shipping papers and their varying formats make it difficult for responders to quickly determine the contents of vehicles. In many cases the delays are costly, especially in terms of inefficient use of response resources. The frequency of such problems in the case studies suggests that noncompliance with regula- tions may contribute to the problem. Missing Emergency Response In formation in Shipping Documents Federal regulations require that emergency response information ac- company or be included on hazardous materials shipping papers. The information required includes a description of immediate health hazards, warnings of the risk of fire and explosion, and preliminary methods of handling fires and spills. Information that must accom- pany shipments includes the proper material name, hazard class, and the shipper's 24-hr emergency telephone number. In several case studies the required information was incorrect or missing, and some of the responders noted that shipping papers do not always contain a working emergency telephone number. Failure to provide a working 24-hr telephone contact is difficult to excuse as a misinterpretation of regulations and should be amenable to reduction through more rigorous enforcement. Inspections Most enforcement efforts are conducted at the state level, including efforts to enforce federal motor carrier requirements. In recent years, states have taken on a larger role in hazardous materials inspection. DOT's Motor Carrier Safety Assistance Program (MCSAP) has been successful in enhancing the quantity and quality of inspections at the state level. MCSAP funding is provided to states to enforce (through inspections) all federal motor carrier safety regulations. However, states have discretion to set enforcement priorities, and
Options To Improve Information: Nontechnological Components 139 enforcement of hazardous materials regulations is not necessarily a high priority. About one violation in seven recorded under MCSAP involves haz- ardous materials. However, few MCSAP inspections involve examina- tion of the cargo of trucks carrying hazardous materials (U.S. Senate. Committee on Commerce, Science, and Transportation. 1991). Conse- quently, inspectors are unable to verify that containers and packages are labeled and marked properly or confirm that placards and shipping papers are consistent with the cargo. Hazardous materials inspections take longer and require additional inspector training, and these re- source requirements further discourage emphasis on hazardous mate- rials safety in state MCSAP programs. Of course, the states must allocate their enforcement resources to minimize overall highway safety risks. It is not known whether divert- ing resources to hazardous materials enforcement would increase or decrease overall safety. However, without efforts to increase emphasis within MCSAP on hazardous materials or to undertake independent enforcement activity, it is less likely that states will increase their enforcement efforts in this area. Filing of Complaints by Emergency Responders If violations of hazardous materials transportation regulations fre- quently aggravated the consequences of incidents, such violations would probably be witnessed often by emergency responders. How- ever, since enforcement is not the responder's primary concern at an incident, such observations often go unreported. Actions to encourage and facilitate the filing of federal and state complaints by responders who witness violations might help to penalize some serious offenses. A pilot project to establish links between one or more metropolitan fire departments and state hazardous materials enforcement officials for reporting suspected violations could be a starting point. TRAINING OPTIONS Two categories of information problems identified in the case studies concerned inadequate emergency actions on the part of public emer- gency responders and vehicle operators (i.e., truck drivers and train
140 HAZARDOUS MATERIALS SHIPMENT INFORMATION crews). In several case studies, for instance, emergency responders were not aware of all information sources available to them (e.g., shipper emergency telephone number and CHEMTREC) or were not prepared to take advantage of the sources. Some vehicle operators performed inadequatelyâfor instance, by leaving the scene of the accident with- out contacting responders or by failing to retrieve the shipping docu- ments from the vehicle when they were readily available. Hence, in addition to improving regulations and enforcement, training and other preparation of response personnel and vehicle operators must be considered. Responder Training Training of responders is necessary to ensure that information sources are used as intended. NTSB has made several recommendations con- cerning needed improvements in responder training for hazardous materials incidents (NTSB Recommendations R-90-040, R-88-055, and 1-83-001; see Appendix Q. Most responders do not need the comprehensive hazardous materials training that might be appropriate for members of special hazardous materials response units. First re- sponders need to understand the importance of locating vehicle opera- tors to obtain documents and other information and to be aware of available reference sources (e.g., shipper and CHEMTREC), and they should have basic operational training that allows them to quickly locate and interpret hazard information from available sources (e.g., placards). Congress acknowledged the importance of training in HMTUSA by establishing an emergency response training and grants program. States will be paid up to 80 percent of the cost incurred for training and planning for hazardous materials incidents. Funding will be generated by a fee assessed on the hazardous materials industry. More than 30,000 volunteer and professional fire departments are in operation in the United States, with 1.5 million fire fighters and emergency medical response personnel. Funds generated by the new federal fee are ex- pected to be small compared with the annual cost of training emer- gency responders to deal with hazardous materials or with training costs arising directly from federal requirements concerning emergency preparedness of state and local governments.
Options To Improve Information: Nontechnological Components 141 Forethought is needed to ensure that the federal funds provided by HMTUSA are used productively. Under the current program, most of the funds will be allocated to states through grants. States will be able to use the grants to supplement or develop training programs. The relatively small amount of funds (when divided among SO states) suggests that voluntary grant use guidelines by DOT would be helpful. A portion of the federal funding for training is being used by the Research and Special Programs Administration (RSPA) to develop course curriculum for hazardous materials response. Given the limited funds, a curriculum aimed at improving first responder awareness training programs might be suitable. At the state level, the appropriate use of funds will vary according to individual state needs. Guidance by DOT in the form of exemplary or model training activities might help the states in the use of limited funds. OSHA regulates the protection of workers engaged in hazardous materials operations and accredits training programs for workers covered under the regulations. OSHA has issued proposed accredita- tion regulations (55FR2776, Jan. 26, 1990; 57FR4858, Feb. 10, 1990). The proposed regulations do not cover emergency re- sponders. A study of emergency responder training conducted for OSHA concluded that only 35 percent of emergency response courses met five minimum characteristics defined by the study, and many did not meet the OSHA-established minimum hour require- ments (Eastern Research Group n.d.). Specific Training Program on Hazard Identification It would be appropriate to include in responder training a specific course related to hazardous materials identification and communica- tion. The course would be directed toward helping responders locate and interpret hazard information quickly. DOT's hazardous materials communication rules would be thoroughly treated. The training would be directed entirely toward obtaining and interpreting hazardous mate- rials information in a critical transportation emergency; it would not be an adjunct to other training related to response procedures. Responders would be taught about applicable DOT regulations, related EPA and OSHA rules, and sources of shipper, carrier, chemi- cal, emergency response, and other data bases. Numerous such data bases exist that could be useful if understood and made more easily accessible.
142 HAZARDOUS MATERIALS SHIPMENT INFORMATION Cost and Delivery Mechanisms Planning of training needs should begin with study of the costs and delivery of training to hazardous materials emergency responders and enforcement officers. It would be appropriate for DOT and the U.S. Fire Administration of the Federal Emergency Management Agency to conduct such a study jointly. Participation of EPA and the National Institute of Environmental Health Sciences would also be valuable. The study should be undertaken with the oversight.of Congress. Its product should be a report to Congress on national requirements and the federal role in training. The main objective of the study would be to examine the organiza- tion of training delivery nationwide and current and required funding levels, not to develop or define curriculum content. The study should identify the most cost-effective means of delivering training, using existing courses and the existing network of qualified instructors. The study should also identify the full costs associated with meeting federal mandates in support of public-sector emergency response training. The study should consider the advantages and disadvantages of financing through user fees and the trade-offs between reliance on a single national fee and the multiple state fees that are now being enacted. The study could, also examine matching requirements for existing federal grant funds available for emergency response training and eligibility of spending for training equipment under current federal grant programs. The study should consider mechanisms for providing consultation between DOT and state and local governments on administration of grant programs and other training initiatives. Participation and input of state and local emergency responders is essential for success of the study. Often these agencies lack the awareness and funds necessary to respond to invitations to contribute to or comment on such undertak- ings. DOT should take whatever steps are necessary to ensure adequate state and local involvement. Emergency Training for Transportation Workers In the case studies in which information failures did not occur, the vehicle operator (truck driver or train crew) was often a critical source of information to responders. As required by HMTUSA, DOT has
Options To Improve Information: Nontechnological Components 143 recently enacted regulations concerning minimum training standards for transportation workers (Docket HM-126F). DOT's rules require that first response training be provided to transport workers who could be involved in a hazardous materials incident. [OSHA is responsible for establishing training requirements for workers whose primary respon- sibility is emergency response (e.g., training of carrier and shipper emergency response teams).] The rules allow companies to develop and implement their own training programs. The companies must test each employee to ensure that the training received is effective. The rules call for various levels of training, depending on the employee's role in handling hazardous materials. All employees involved with hazardous materials must receive general awareness training in hazardous mate- rials communications requirements (e.g., placards), whereas train crews and truck drivers must be provided with more detailed safety instruction that describes their duties and responsibilities in the event of an incident. NTSB has issued numerous recommendations concerning the need for carriers to train their vehicle operators and supervisors on the actions they should take immediately following an accident (NTSB Recommendations R-90-039, R-90-028, R-87-056, R-87-055, and I-85-006). The recommendations have generally advised carriers to train operators and their supervisors to immediately notify emergency response personnel in the event of an incident, to identify themselves to emergency response personnel at the scene, and to provide on-board emergency documentation if possible. The training is especially impor- tant for train crews, who may be some distance from the incident if it occurs at the middle or rear of the train. In such cases, it is vital that train crews locate the appropriate response personnel and provide them with the proper on-board documentation. NTSB recently com- pleted a study of 45 rail incidents involving hazardous materials. It found fault with the manner in which several railroads provide infor- mation and training on hazardous materials to their own employees (NTSB 1991). Most of the training is provided through classroom instruction. NTSB suggested that the training be supplemented by more in-depth training on the purpose of emergency response informa- tion and how to provide it to response authorities in an emergency. The new RSPA rules on hazardous materials transport worker train- ing do not provide guidelines on what elements a training program should emphasize. Writing comprehensive guidelines would be diffi- cult because the duties of affected workers are extremely diverse. How-
144 HAZARDOUS MATERIALS SHIPMENT INFORMATION ever, NTSB has identified problems that RSPA should require railroads and other carriers to cover. Many hazardous materials carriers and shippers have developed successful training programs for their em- ployees on their own initiative. It may be beneficial for DOT to evaluate some of these programs to determine which elements have contributed to their success. In this way DOT could provide guidelines to carriers and shippers about areas that must be covered in training (e.g., opera- tor knowledge of the importance of providing shipping papers to local responders) and possible mechanisms (e.g., instructional methods) for providing it. This would allow carriers and shippers to maintain the flexibility to develop their own training systems while ensuring that essential training needs are met. Training of Hazardous Materials Inspectors The roadside truck safety inspection data suggest that citation practices vary among the states, and industry has complained of inconsistent enforcement. Lack of consistency can seriously impair the effectiveness of enforcement efforts. DOT's compliance monitoring should include comparison of enforcement practices among jurisdictions. On the basis of the monitoring, DOT could provide guidance or assistance to the enforcement agencies to ensure uniform standards for training of in- spectors in hazardous materials regulations and related inspection and citation procedures. Emergency Preparation Preparation for hazardous materials transport incidents is more diffi- cult than for fixed-site incidents. Planning for hazards at fixed sites, such as chemical plants, can be coordinated between local responders and facility operators. Mock drills can be conducted, emergency tele- phone numbers readily updated, and determinations made about man- power and equipment needs in the event of an accident. Transportation incidents, on the other hand, can occur where trained personnel and proper equipment are not available and responders and transporters have not had the opportunity to make preparations. Railroad yards or depots, truck terminals, and loading docks of large shippers, which are essentially fixed-site facilities, are exceptions. To a certain extent, pre-
Options To Improve Information: Nontechnological Components 145 dictable commodity movements on railroad tracks and regular truck- ing routes allow carriers and communities to make preparations. The NTSB study on hazardous materials rail incidents found nu- merous cases of poor planning and coordination between railroads and local communities (NTSB 1991). The study found that in some cases rail carriers and local responders did not even have a listing of each other's emergency telephone numbers. Many had not conducted joint exercises or made any other preparations, such as initial notification procedures and written and tested emergency response plans. NTSB recommended that major railroads, in coordination with communities adjacent to railroad yards and along routes, develop and keep current written emergency response plans that include a listing of key railroad personnel, identification of technical resources for handling emergen- cies, procedures for coordination of activities between rail and public responders, and a means of testing response capabilities. Some efforts are being made by private industry to rectify these problems. For instance, the Association of American Railroads and the Chemical Manufacturers Association have established an Interindustry Task Force that has identified "key routes" where railroads will focus their efforts on training and emergency planning. This may improve the situation in many communities if the information is kept current. During the past S years, the Chemical Manufacturers Association has developed a program, called TransCAER, that encourages hazardous materials transportation companies to contact local community emer- gency planners to inform them about the hazardous materials being transported through their community. The goal of the program is to improve coordination. As a minimum, emergency telephone numbers are to be maintained. The Superfund Amendments and Reauthoriza- tion Act of 1986 requires the development of Local Emergency Plan- ning Committees to develop plans for chemical accidents. Industry is required to participate in the planning; however, transporters are ex- empt. In some cases the Local Emergency Planning Committees are being used by transportation companies as a central point for up-to- date information on local response capabilities and emergency tele- phone numbers (TRB 1988). The training program established under HMTUSA allows state and local governments to use funds for emergency planning and prepara- tion. DOT should study some of the methods being used by local governments and industry to improve emergency planning and provide this information to state and local governments applying for grants.
TABLE 5-1 Potential Regulatory, Enforcement, and Training Solutions to Information Failures Problem Category Specific Problems Possible Means of Alleviation Required sources of information are Placards missing or incorrect Systematically monitor compliance as missing or inaccurate Shipping papers missing, incorrect, or the basis for establishing targets for incomplete compliance improvement and measuring progress toward them Upgrade training of motor vehicle inspectors to ensure more uniform enforcement Improve effectiveness of enforcement of hazardous material identification requirements in vehicle safety inspections Monitor the effectiveness of required shipper and carrier employee training programs Establish higher fines for certain hazardous materials identification violations, especially improper placarding and failure to provide telephone contact Facilitate the filing of complaints by emergency responders who discover violations at incidents
Placards, shipping papers, or other information sources are obscured, destroyed, or inaccessible because of a crash or fire or threat of fire, explosion, or toxic exposure Information sources are in compliance with regulations and accessible, but they fail to efficiently convey important information to responders Placards destroyed or obscured Shipping papers destroyed or inaccessible Dangerous placard not descriptive of hazards Multiple shipping papers that are cumbersome to review Responders did not trust all information sources available and therefore did not use them Improve placard visibility through use of retroreflective materials and other means Evaluate new materials for prolonging the fire resistance of placards Ensure that shipping papers and placards on vehicles carrying mixed loads are comprehensible to emergency responders by (a) requiring a summary manifest listing critical information about each hazardous shipment on board a vehicle, (b) requiring carriers to provide 24-hr telephone contact to a person with access to vehicle cargo information, (c) replacing the Dangerous placard with a more specific message, and (d) changing the minimum quantities requiring placarding Analyze the use of CHEMTREC and the National Response Center by local emergency responders to determine whether needed services are being provided (continued on next page)
TABLE 5-1 (continued) Problem Category Specific Problems Possible Means of Alleviation Coordinate requirements for material identification among DOT, EPA, and 051-IA to ensure that they are consistent and understood by responders; in particular, eliminate ambiguities concerning identification of hazardous contents of trailers or railcars being used as storage Require that a standardized material data sheet accompany shipping papers for certain hazardous materials Information is insufficient because the material or shipment is exempt from some federal hazardous materials transportation regulations Some potentially harmful materials or shipments are not covered by DOT regulations governing placarding, labeling, and shipping papers because they are not designated as Encourage industry efforts to adopt uniform tank trailer numbering
hazardous, do not meet minimum weight/quantity thresholds, or are not involved in a clearly transport- related activity Responders sometimes cannot readily confirm that a material is not hazardous Vehicle operator is unprepared to provide information Responders fail to obtain or use available information Truck driver unable to identify basic hazards of cargo Train crew did not take proper actions to assist responders, such as identifying themselves and providing on-board documentation Responders unaware of all information sources available or unprepared to take advantage of all available information Ensure that training of shipper and carrier personnel covers communication with local emergency responders Expand federal resources and technical assistance for training of responders in the skills of hazard identification and interpretation Develop a hazardous materials identification training curriculum
150 HAZARDOUS MATERIALS SHIPMENT INFORMATION SUMMARY Six categories of failures of the existing information system were identi- fied in Chapter 3. The systematic approach to improving information suggested at the beginning of this chapter would focus on devising a coherent strategy to address each problem. Table 5-1 correlates the possible regulatory, enforcement, and training improvements de- scribed in this chapter with the six categories of failure. Proposals have been made that are relevant to each of the problem categories. Evalua- tion will be necessary to determine whether each proposal is worth- while and whether, taken together, the proposals are sufficient for dealing with the problems. REFERENCES ABBREVLA11ONS NTSB National Transportation Safety Board TRB Transportation Research Board Eastern Research Group (n.d.). Survey of Emergency Response Training Programs. DOL/OSHA Contract J-9-F-8-0035. NTSB. 1991. Safety Study: Transport of Hazardous Material by Rail. Report NTSB/SS-91-01. Washington, D.C. TRB. 1988. Special Report 219: Pipelines and Public Safety. National Research Council, Washington, D.C. U.S. Senate. Committee on Commerce, Science, and Transportation. 1991. Reau- thorization of the Motor Carrier Safety Assistance Program (MCSAP): Options Intended To Improve Highway Safety. March.
APPENDIX F A Sections I and 25 of the Hazardous Materials Transportation Uniform Safety Act of 1990 P ublic Law 101-615, 101st Congress, November 16, 1990 (104 Stat. 3244) An Act To amend the Hazardous Materials Transportation Act to authorize appropriations for fiscal years 1990, 1991, and 1992, and for other purposes. Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled. SECTION 1. SHORT TITLE: REFERENCE: TABLE OF CONThNTS. SHORT Trn.E.âThis Act may be cited as the "Hazardous Mate- rials Transportation Uniform Safety Act of 1990". REFERENCE.âEXCCpt as otherwise specifically provided, when- ever in this Act an amendment or repeal is expressed in terms of an amendment to, or repeal of, a section or other provision, the reference shall be considered to be made to a section or other provision of the Hazardous Materials Transportation Act. TABLE OF CoNTENTS.â Sec. 1. Short title; reference; table of contents. Sec. 2. Findings. 151
152 HAZARDOUS MATERIALS SHIPMENT INFORMATION Sec. 3. Definitions. Sec. 4. Federal regulatkns governing transportation of hazardous materials. Sec. S. Representation and tampering. Sec. 6. Disclosure. Sec. 7. Handling of hazardous materials. Sec. 8. Hazardous materials transportation registration; motor carrier safety permits. Sec. 9. Exemptions. - Sec. 10. Definition of certain materials. Sec. 11. Secretary's powers. Sec. 12. Penalties. Sec. 13. Relationship to other laws. Sec. 14. Funding. Sec. 15. Transportation of certain highly radioactive materials. Sec. 16. Inspectors. Sec. 17. Public sector training and planning. Sec. 18. Hazmat employee training grant program. Sec. 19. Railroad tank cars. Sec. 20. Application of Federal, State, and local law to Federal contractors. Sec. 21. Railroad tank car study. Sec. 22. Uniformity of State motor carrier registration and permitting forms and procedures. Sec. 23. Financial responsibility. Sec. 24. Federally leased commercial motor vehicles. Sec. 25. Improvements to hazardous materials identification systems. Sec. 26. Continually monitored telephone systems. Sec. 27. Shipper responsibility report. Sec. 28. State participation in investigations and surveillance. Sec. 29. Retention of markings and placards. Sec. 30. Relationship to Federal Railroad Safety Act of 1970. Sec. 31. Effective date. SEC. 25. IMPROVEMENTS TO HAZARDOUS MATERIALS IDENTIFICATION SYSTEMS (a) RULEMAKING PROCEEDING.- (1) INmATION.-In order to develop methods of improving the current system of identifying hazardous materials being trans-
Hazardous Materials Transportation Uniform Safety Act of 1990 153 ported in vehicles for safeguarding the health and safety of per- Sons responding to emergencies involving such hazardous mate- rials and the public and to facilitate the review and reporting process required by subsection (d), the Secretary of Transporta- tion shall initiate a rulemaking proceeding not later than 30 days after the date of the enactment of this Act. (2) PRIMARY PuRPosEs.âThe primary purposes of the rulemak- ing proceeding initiated under this subsection areâ to determine methods of improving the current system of placarding vehicles transporting hazardous materials; and to determine methods for establishing and operating a central reporting system and computerized telecommunica- tions data center described in subsection (b)(1). (3) METHODS OF IMPROVING PLACARDING sysmi.âThe methods of improving the current system of placarding to be considered under the rulemaking proceeding initiated under this subsection shall include methods to make such placards more visible, methods to reduce the number of improper and missing placards, alternative methods of marking vehicles for the purpose of identi- fying the hazardous materials being transported, methods of modifying the composition of placards in order to ensure their resistance to flammability, methods of improving the coding system used with respect to such placards, identification of ap- propriate emergency response procedures through symbols on placards, and whether or not telephone numbers of any continu- ally monitored telephone systems which are established under the Hazardous Materials Transportation Act are displayed on vehicles transporting hazardous materials. (4) COMPLETION OF RULEMAKING PROCEEDING WITH RESPECF TO REPORTING SYSTEM AND DATA CENTER.âNot later than 19 months after the date of the enactment of this Act, the Secretary of Trans- portation shall complete the rulemaking proceeding initiated with respect to the central reporting system and computerized telecom- munications data center described in subsection (b). (5) FINAL RULE WITH RESPECT TO PLACARDING.âNot later than 30 months after the date of the enactment of this Act, the Secretary of Transportation shall issue a final rule relating to improving the current system for placarding vehicles transporting hazardous materials.
154 HAZARDOUS MATERIALS SHIPMENT INFORMATION (b) CENTRAL REPORTING SYSTEM AND COMPUTERIZED TELECOMMUNICA- liONS DATA CENTER STuDy.â (1) ARRANGEMENTS WITH NATIONAL ACADEMY OF SCIENCES.âNot later than 30 days after the date of the enactment of this Act, the Secretary of Transportation shall undertake to enter into appro- priate arrangements with the National Academy of Sciences to conduct a study of the feasibility and necessity of establishing and operating a central reporting system and computerized tele- communications data center that is capable of receiving, storing, and retrieving data concerning all daily shipments of hazardous materials, that can identify hazardous materials being trans- ported by any mode of transportation, and that can provide information to facilitate responses to accidents and incidents involving the transportation of hazardous materials. (2) CONSULTATION AND REPORT.âIn entering into any arrange- ments with the National Academy of Sciences for conducting the study under this section, the Secretary of Transportation shall request the National Academy of Sciencesâ to consult with the Department of Transportation, the Department of Health and Human Services, the Environ- mental Protection Agency, the Federal Emergency Manage- ment Agency, and the Occupational Safety and Health Administration, shippers and carriers of hazardous mate- rials, manufacturers of computerized telecommunications systems, State and local emergency preparedness organiza- tions (including law enforcement and firefighting organi- zations), and appropriate international organizations in conducting such study; and to submit, not later than 19 months after the date of the enactment of this Act, to the Secretary, the Committee on Commerce, Science, and Transportation of the Senate, and the Committees on Energy and Commerce and Public Works and Transportation of the House of Representatives a report on the results of such study. Such report shall include recommendations of the National Academy of Sciences with respect to establishment and operation of a central reporting system and computerized telecommunica- tions data center described in paragraph (1). (3) AUTHORIZATION OF APPROPRIATION.âJn addition to amounts authorized under section 115 of the Hazardous Materials Trans-
Hazardous Materials Transportation Uniform Safety Act of 1990 155 portation Act, there is authorized to be appropriated to the Secretary of Transportation to carry out this subsection $350,000. (c) ADDITIONAL PURPOSES OF RULEMAKING PROCEEDING AND STuDY.â Additional purposes of the rulemaking proceeding initiated under subsection (a) with respect to a central reporting system and comput- erized telecommunications data center described in subsection (b) and the study conducted under subsection (b) areâ to determine whether such a system and center should be established and operated by the United States Government or by a private entity, either on its own initiative or under contract with the United States; to determine, on an annualized basis, the estimated cost for establishing, operating, and maintaining such a system and cen- ter and for carrier and shipper compliance with such a system; to determine methods for financing the cost of establishing, operating, and maintaining such a system and center; to determine projected safety benefits of establishing and operating such a system and center; (S) to determine whether or not shippers, carriers, and han- dlers of hazardous materials, in addition to law enforcement officials and persons responsible for responding to emergencies involving hazardous materials, should have.access to such system for obtaining information concerning shipments of hazardous materials and technical and other information and advice with respect to such emergencies; to determine methods for ensuring the security of the information and data stored in such a system; to determine types of hazardous materials and types of shipments for which information and data should be stored in such a system; to determine the degree of liability of the operator of such a system and center for providing incorrect, false, or misleading information; to determine deadlines by which shippers, carriers, and handlers of hazardous materials should be required to submit information to the operator of such a system and center and minimum standards relating to the form and contents of such information;
156 HAZARDOUS MATERIALS SHIPMENT INFORMATION (10) to determine measures (including the imposition of civil and criminal penalties) for ensuring compliance with the dead- lines and standards referred to in paragraph (9); and (11), to determine methods for accessing such a system through mobile satellite service or other technologies having the capa- bility to provide 2-way voice, data, or facsimile services. (d) REVIEW AND REPORT TO CONGRESS.â IN GENERAL.âNot later than 25 months after the date of the enactment of this Act, the Secretary of Transportation shall re- view the report of the National Academy of Sciences submitted under subsection (b) and the results of rulemaking proceeding initiated under subsection (a) with respect to a central reporting system and computerized telecommunications data center and shall prepare and submit to Congress a report summarizing the report of the National Academy of Sciences and the results of such rulemaking proceeding, together with the Secretary's rec- ommendations concerning the establishment and operation of such a system and center and the Secretary's recommendations concerning implementation of the recommendations contained in the report of the National Academy of Sciences. WEIGHT TO BE GIVEN TO RECOMMENDATIONS OF NAS.âIn conducting the review and preparing the report under this sub- section, the Secretary shall give substantial weight to the recom- mendations contained in the report of the National Academy of Sciences submitted under subsection (b). INCLUSION OF REASONS FOR NOT FOLLOWING RECOMMENDA- TIONS.âIf the Secretary does not include in the report prepared for submission to Congress under this subsection a recommenda- tion for implementation of a recommendation contained in the report of the National Academy of Sciences submitted under subsection (b), the Secretary shall include in the report to Con- gress under this subsection the Secretary's reasons for not recom- mending implementation of the recommendation of the National Academy of Sciences.
APPENDIX 1~ Data Needs and Sources The committee's efforts to evaluate the design and performance of the existing information system' were hampered by inadequate data and the lack of previous studies on the subject. The committee had to devote much of its time and resources to gathering and examining very limited data to make rough determinations about the performance of the existing system and to identify problem areas. The purpose of this appendix is to describe the types of data the committee needed to conduct the study, provide suggestions for improving available data, and describe the data sources ultimately used. DATA NEEDS The kinds of data the committee would have liked to have had and believes are critical for a more substantive evaluation would address the following questions: To what degree are shippers and carriers complying with U.S. Department of Transportation (DOT) regulations governing emer- gency information, such as placarding and shipping paper require- ments, and how can compliance be improved? 157
158 HAZARDOUS MATERIALS SHIPMENT INFORMATION How effective are the federal regulations in meeting the informa- tion needs of responders to hazardous materials incidents, and how can their effectiveness be improved? How effective are other commonly used information resources such as CHEMTREC and Material Safety Data Sheets (MSDSs), and how can their effectiveness and availability be improved? f To what degree are responders using existing information sources? What are the training needs of emergency responders, enforce- ment officials, and shippers and carriers of hazardous materials to ensure that emergency information is provided and properly used? To help answer these questions, objective evidence on the perfor- mance of the existing information system is required. The principal source of data used by DOT to evaluate the safety performance of hazardous materials transportation is the Hazardous Materials Infor- mation System (HMIS), which contains reports on thousands of haz- ardous materials incidents filed by carriers. HivilS was designed to determine the performance of hazardous materials packages and con- tainers, which is a regulatory and enforcement responsibility of the Research and Special Programs Administration (RSPA). However, HMIS was not designed to determine the performance of placards, shipping papers, and other required information sources and is not well suited for this purpose, because HMIS incident reports are filed by carriers rather than emergency responders. The best source of performance data is the National Transportation Safety Board (NTSB). As a standard practice, NTSB examines the adequacy of emergency response operations during its investigations of major hazardous materials incidents. The success of past NTSB investi- gations in uncovering emergency information problems suggests that a promising way to study the performance of the information system is through similar detailed examination of actual incidents while focusing on emergency response. Such an approach might consist of a systematic survey of incidents responded to by state and local police and fire departments. Follow-up interviews would be conducted with the re- sponders to determine the performance of information sources. Similar data programs already exist within DOT for other purposes. An example is the National Accident Sampling System (NASS) of the National Highway Traffic Safety Administration (NHTSA), which consists of detailed investigations of accident causes and circumstances using a statistically valid sample of motor vehicle accidents drawn from
Data Needs and Sources 159 state and local police accident reports. NASS sampling techniques allow extrapolation of the results to develop national estimates with far more detail than otherwise would be possible from a national census of incidents. To aid in monitoring the performance of its hazardous materials program, DOT could also make systematic use of existing data bases that contain relevant information. Examples are the Federal Highway Administration's Safetynet System, NFITSA's NASS and Fatal Acci- dent Reporting System, the University of Michigan's data base on Trucks Involved in Fatal Accidents, the U.S. Fire Administration's National Fire Information Reporting System, the Environmental Pro- tection Agency's (EPA's) Acute Hazardous Events data base, CHEM- TREC's incident notification records, and various state hazardous materials incident records [e.g., the California Hazardous Materials Incident Reporting System (CI-IMIRS)]. Use of the data and improvements in data compatibility could be aided by the development by DOT of an ongoing relationship with other federal and state agencies with data programs, such as the U.S. Fire Administration. For instance, a hazardous materials data stan- dards committee might be established to foster voluntary standardiza- tion of data. The goal of such a committee would be to develop, to the extent practical, common data elements, standard data element def- initions, more uniform incident reporting criteria, and consolidated reports from the various data systems. Potential benefits of these im- provements would be a larger and more reliable pool of incidents to sample for detailed case study examinations, reductions in duplicate reporting, greater ability to verify and improve the accuracy of individ- ual data bases, and national statistics for use in determining trends in hazardous materials safety (e.g., kinds of chemicals involved, common location of incidents, and year-to-year changes in incidents) and in identifying the changing training and equipment needs of emergency responders. CASE STUDIES The limited availability of performance data compelled the committee to conduct its own case study investigations in order to provide an objective means of evaluating the existing information system. A total of 125 incident cases were examined, including 80 truck and 45 rail
160 HAZARDOUS MATEIUALS SHIPMENT INFORMATION cases. All had severe, or consequential, outcomes. Combined, the cases involved 260 injuries (fatal and nonfatal), 49 workplace and public evacuations, 27 fires and explosions, and 23 major highway closings. Consequential cases were intentionally selected for examination be- cause they tend to be well documented and are of most concern to responders and the public. Severe consequences are not typical of hazardous materials inci- dents, which seldom involve major wrecks, fires, injuries, public evac- uations, or other serious outcomes. Most of the adverse consequences were not caused by problems with emergency response but were the direct result of the crash, derailment, or other event causing the incident. The cases were derived from three sources, which are described in the following subsections. California Truck Cases Forty hazardous materials truck incidents that occurred in California during 1990 and the first half of 1991 were examined to determine whether emergency information sources performed adequately. The incidents were selected because they involved an injury (fatal or non- fatal), evacuation, or major highway closing. Such consequential in- cidents were examined to ensure that responders would recall the incidents. The incidents were identified from two data bases: CHMIRS and RSPA's HMIS. California was chosen as the location for the case studies because the combined CHMIRS and RSPA data provide a large source of truck incident records from which to draw (more than 1,500 reported incidents to both data bases during the 18-month period). State and local emergency responders in California are required to report all hazardous materials incidents to CHMIRS. CHMIRS reports contain general information on the type of haz- ardous materials incident and its consequences. CHMIRS reports from 1990 (1991 CHMIRS data were not available, though RSPA data were) were obtained for 19 incidents that met the severity consequences described earlier. The incidents consisted of all consequential incidents reported, accounting for about 4 percent of total truck incidents in CHMIRS. Newspaper reports were reviewed, and the responders sub- mitting the report for each incident were contacted by telephone and asked to describe the incident response and to answer questions about
Data Needs and Sources 161 the availability, accuracy, and timeliness of emergency information. Cases were classified as having information problems if one or more of the critical kinds of information discussed in Chapter 3 were delayed, inaccurate, missing, or not properly consulted by responders. Five cases had problems involving placards, shipping papers, or other emergency information sources. A statistical summary of the CFLMIRS cases is provided in Table B-i (cases numbered 1 to 19). An additional 21 RSPA-reported incidents were examined that oc- curred in California during 1990 and the first half of 1991. Again, only consequential incidents were examined. Cases were selected on the basis of information found in the HMIS report. In each case newspaper reports were reviewed and the parties involved in the incident were contacted to determine whether information problems were encoun- tered during the response. Responders encountered information prob- lems in three. A statistical summary of the RSPA cases is provided in Table B-i (cases numbered 20 to 40). A total of 40 California truck incidents were examined. In 8 (20 percent), responders encountered information problems that adversely affected the response. The following is a summary of the problems encountered in the eight cases (numbered according to their order in Table B-i): 3. Two workers at a truck terminal were overcome by fumes while unloading a tractor-semitrailer. The trailer, which was not placarded, contained several shipments. The terminal manager reviewed the ship- ping papers and could not find hazardous materials listed. Eventually he identified the leaking material as a herbicide contained in several 5-gal cans. The shipping papers did not contain information on the chemical composition or potential hazards of the herbicide. Concerned that the herbicide might be hazardous, the terminal manager called local fire officials for assistance. Fire officials contacted the shipper, who claimed the shipment was not hazardous and therefore did not require placards or more descriptive shipping papers. The shipper faxed the MSDS to fire officials. Because of the fumes coming from the trailer, fire officials and the trucking company treated the herbicide as hazardous and hired a private disposal company to assist with cleanup. 7. In a rural farming area, a stationary trailer containing several drums of pesticides spontaneously ignited. Fire fighters responding to the fire were not aware of the pesticides, which were stored illegally. The trailer was not placarded. Fire fighters initially attempted to extin-
TABLE B-i Summary of 40 California Hazardous Materials Truck Cases Truck and Trailer Type Consequences Combi- Cargo Hazard Class ILA Information nation Straight j7 En Van! Van! Comb. Case Route Flat Tank Flat Tank None Mat. 1 X 2 X 3 X 4 X S X x x 6 X 7 X 8 X X X 9 X 10 X X 11 X X x 12 X X X 13 X X 14 X X X is x x x 16 X X X 17 X X 18 X 19 X X 20 X 21 X X 22 X 23 X No. of ajor Hwy. Problem Areas No. of No. of People Closed Fire! Spill/ Shipping Corr. Poison Other Fatalities Injuries Evacuated (hr) Expi. Vapor Placard Papers Other x 1 x X 2 X X 2 X X 2 X 1 S x 1 x X 2 20 X b 1 6 X x 1 x x 1 x 4 X 1 20 14 X X X 1 8 X 1 8 X 1 2 50 12 X X 1 x X 1 14 X a a x 1 x 2 X x 1 x 1 x x 1 x x 1. x
164 HAZARDOUS MATERIALS SHIPMENT INFORMATION guish the fire, but after the owner of the trailer, who arrived 1/2 hr later, reported the presence of the pesticides, the decision was made to let the fire burn. Twenty residents in the path of the smoke were evacuated for a short time. Two residents were taken to the hospital for evaluation after complaining of smoke inhalation. 13. The driver of a tractor-semitrailer was briefly overcome by fumes while stopped on the roadside investigating a rattling sound in the trailer. A passing police officer stopped to assist the driver. The officer noticed corrosive placards on the vehicle and alerted local fire officials. The truck driver, who could only identify the load in general terms as swimming pool chemicals, presented the shipping papers to the officer and fire fighters. The shipping papers were in disorder and confusing. The receiver of the shipment, who was located nearby, was contacted by fire officials to help identify the shipment. Identification took about 30 mm. No one was injured in the incident except for the driver, who received minor injuries. The roadway was closed for 8 hr during the response. The driver of a tractor-semitrailer was overcome by fumes while inspecting the vehicle on the roadside. Passing police officers noticed the injured driver and applied first aid after moving him away from the scene. Corrosive placards were observed by the officers, who alerted fire units. The shipping papers were retrieved from the truck cab and reviewed by fire officials. The papers identified the materials and their quantities and packaging. Fire officials, however, could not determine which packages were leaking in the enclosed van. Concerned about a chemical reaction, the fire fighters closed the highway and began isolating the area until a private hazmat team arrived 4 hr later. The fire fighters did not attempt to call the shipper (whose telephone number was on the shipping papers) until after the hazmat team had arrived. CHEMTREC was not called. The highway was closed for 14 hr during the response. An employee of a landfill firm, who was inspecting a truckload for disposal, was overcome by fumes. The operator of the landfill suspected that the load was an illegal shipment of hazardous waste and notified local fire officials. The truck, which was not placarded, was found to contain 25 drums of hazardous waste. County health inspec- tors were called to investigate. The injured worker was taken to a hospital for evaluation and later returned to work. 30. A forklift operator punctured a drum of potassium hydroxide while loading it onto a trailer at a truck terminal. About 40 gal of liquid
24 X X 8 X 25 X X 95 X 26 X X X I X 27 X X X 1 X 28 X X 2 X 29 X X X 3 300 X X 30 X X S x 31 X X 30 X 32 X X X 12 X 33 X X X X X 100 10 X h 34 X X 1 3 X X 35 X X I X 36 X X 1 X 37 X X X 2 X 38 X X 20 X 39 X X X 100 X 40 X X 12 X Total 17 17 14 2 3 4 19 11 7 S 2 45 760 91 5 38 0 6 5 'Herbicide not designated as a hazardous material by DOT; accordingly, the vehicle was not placarded, and the shipping papers and package labels did not contain emergency information needed by responders. The shipper was eventually contacted and provided MSDS. bA storage trailer containing various pesticides caught fire. The trailer was not placarded. The pesticides were being stored in violation of local ordinance. Fire fighters were not aware of the materials until 30 miii after they began extinguishing the fire. Evacuation of nearby residents in the path of the smoke was delayed as a result. 'Truck driver could identify load in general terms only. Shipping papers were in disorder. Load consisted of several hazardous cargoes, which caused responders to be concerned about the potential for a chemical reaction. "shipping papers were cumbersome to review because of mixed load. Responders did not try to contact shipper for 4 hr, although telephone number was available. 'Illegal shipment of hazardous waste. As a result, the truck did not have placards or shipping papers. 'Shipping papers for a nonhazardous shipment in a mixed load were not accurate, leading to concern that the shipment could be hazardous and react with a spilled hazardous material. 8Shipping papers did not adequately identify hazards of shipment, particularly in relation to shipment size. "Mixed load of flammable liquids, oxidizers, and poisons. Because the format of shipping papers varied by shipment, fire fighters found them cumbersome and time- consuming to review.
Data Needs and Sources 165 spilled onto the trailer floor, contaminating several shipments in the vehicle. The shipping papers for the punctured drum and the contami- nated shipments were retrieved from the terminal office. The material in the drum was identified immediately; however, the shipping papers for one of the contaminated shipments were incomplete. Concerned that the contaminated shipment might be hazardous and react with the spilled potassium hydroxide, the terminal manager called CHEM- TREC and the local fire department. The shipper of the contaminated shipment was contacted to obtain accurate shipping papers. Approx- imately 45 min after the spill, the correct papers were obtained, and it was determined that the contaminated shipment was not hazardous. 31. Upon opening the door of a trailer readied for unloading, termi- nal workers noticed a strong odor and found three cases of 1-gal cans marked "poison" overturned on the trailer floor. The terminal manager obtained the shipping papers for the cases and immediately identified the material and its quantity. The terminal manager contacted CHEM- TREC, which alerted the local fire department. The terminal manager and responding fire fighters did not find the shipping papers adequate to decide on a proper response, although they were in order. The DOT Emergency Response Guidebook was consulted, but given the small quantities involved, the advice was considered inappropriate and ex- treme. Instead, the shipper was contacted and asked to provide the MSDS for the shipment. The MSDS, which took approximately 25 mm to obtain from the shipper, provided sufficient information to responders. 33. The driver of a tractor-trailer van carrying a mixed load of oxidizers, flammable liquids, and poisons stopped the vehicle on the roadside after noticing a leak and smoke coming from the rear of the trailer. The driver, who alerted local fire officials, could not identify the materials in the load but reported the presence of placards. Responding fire fighters were able to identify the materials from the shipping papers, but the process was cumbersome because of the number of papers involved. The fire fighters could not determine which materials in the van trailer were leaking and were concerned about a chemical reaction. CHEMTREC, which was called for advice, provided the telephone numbers of shippers, although this information was pro- vided in the shipping papers. After several hours a private hazmat company was called to handle the incident, at which point the shipper was contacted for advice. The highway was closed for 14 hr.
166 HAZARDOUS MATERIALS SHIPMENT INFORMATION NTSB Truck Cases The California truck cases were supplemented with 40 truck cases selected from a special NTSB study of 189 truck crashes (NTSB 1988). The 40 cases consist of all crashes investigated in the NTSB study that involved trucks carrying hazardous cargoes. They were investigated by NTSB to identify recurring accident causes. NTSB routinely assesses the performance of emergency response operations during its investiga- tions. Investigators made personal visits to the crash site and inter- viewed the carriers and responders involved. Hence, for each case, the committee was able to review the results of the NTSB investigation (from backup documentation on microfiche) pertaining to emergency response operations. The documentation was sufficient to identify cases in which emergency responders encountered information prob- lems and to determine the reported consequences of the problems. A statistical summary of the 40 NTSB truck cases is provided in Table B-2. The NTSB cases tend to have more severe outcomes than the California cases because the NTSB study focused on truck crashes. In 11 of the NTSB cases it appears that responders encountered diffi- culties in obtaining emergency information. The following is a sum- mary of each of the 11 cases (numbered according to their order in Table B-2): 1. A combination unit carrying electrical transformers containing various levels of polychiorinated biphenyls (PCBs) crashed into a pickup truck on an Interstate highway. The truck jackknifed and rolled over, resulting in the ejection of one of the transformers and the release of toxic material onto the highway. The truck was placarded, but the shipping papers were burned in the ensuing fire. Responding fire fight- ers were able to determine the presence of PCBs and the concentration from markings and labels on the transformers. However, some initial confusion was caused by the truck driver, who claimed that no haz- ardous materials were present. 4. The driver of a combination tank truck carrying diphenylmethan disocyanate lost control of the vehicle while crossing an icy section of highway. The truck struck a pedestrian and overturned. Approx- imately 1,200 gal of the cargo spilled. The trailer was not placarded, and the shipping papers were destroyed in the wreck. The driver, who was injured slightly, misidentified the contents of the load, claiming that the load was more hazardous than it actually was. Fire fighters
TABLE B-2 (continued) Consequences Combi- Cargo Hazard Class M Information nation Straight ajor Flam./ No. of Hwy. Problem Areas NTSB Case Van! Van! Comb. No. of No. of People Closed Fire/ Spill/ Shipping Case Number Flat Tank Flat Tank Mat. Corr. Poison Other Fatalities Injuries Evacuated (hr) ExpI. Vapor Placard Papers Other Truck and Trailer Type 23 DEN-86-H-TRO6 X X 24 SEA-87-H-TRO8 X X 25 CHI-87-H-TRO6 X X 1 26 DEN-87-H-TRO7 X X 1 27 FTW-87-H-TRO8 X X I 28 ATL-87-H-TRO6 X X 1 29 DEN-87-H-TR13 X X X 30 DEN-87-H-TR18 X X 31 MKC-87-H-TRO9 X X 3 32 ATL-87-H-TRI3 X X 1 33 DEN-87-H-TR16 X X 34 FFW-87-H-TRO7 X X 35 SEA-87-H-TRI0 X X 36 LAX-87-H-TR13 X X I 37 ATL-87-H-TR18 X X 1 38 DEN-87-H-TR20 X X 39 ATL-87-H-TR19 X 40 LAX-87-H-TRO9 X X Total 10 27 1 2 25 6 4 7 18 x 12 X X x 8 X X 1 10 X X x x x 3 X X 1 x 1 300 24 X 1 6 X X k 1 x 1 x x 3 X 15 x 85 4,525 158 14 31 6 7 7
TABLE B-2 Summary of 40 NTSB Hazardous Materials Truck Cases (NTSB 1988) Truck and Trailer Type Consequences Combi- Cargo Hazard Class M ajor Information nation Straight Flam./ No. of Hwy. Problem Areas NTSB Case Van/ Van/ Comb. No. of No. of People Closed Fire/ Spill/ Shipping Case Number Flat Tank Flat Tank Mat. Corr. Poison Other Fatalities Injuries Evacuated (hr) Expl. Vapor Placard Papers Other 1 FTW-86-H-TR13 X X 1 1,000 iS X X 2 ATL-86-H-TRO8 X X 1 3 3 CHI-86-H-TRO6 X X 1 4 CHI-86-H-TRO8 X X 2 X b b 6 5 ATL-86-H-TRO9 X X 16 2,200 30 X 6 FTW-86-H-TROS X X X 1 7 NYC-86-H-TRO6 X X 1 x d d 8 FTW-86-H-TRO6 X X 1 9 MKC-86-H-TRO6 X X 2 X 10 MKC-86-H-TRO9 X X 18 11 ATL-86-H-TR12 X X 1 2 X 12 FTW-86-H-TR14 X X 4 25 X 13 DEN-87-H-TRO2 X X 3 2 X X 14 LAX-87-H-TRO2 x X 10 X g S 15 MKC-87-H-TRO2 X X X 1 16 ATL-87-H-TRO4 X X 1 22 X 17 NYC-86-H-TRO3 X X 1 X 18 NYC-86-H-TRO5 X X 1 X X 19 LAX-86-H-TR15 X X 1,000 6 X X 20 F1W-87-H-TRO4 X X 2 X 21 FTW-87-H-TRO6 X X 1 X X 22 ATL-87-H-TR24 X X 1 3 10 X X (continued on next page)
Shipping papers burned. Truck driver misidentified material concentration to fire fighters. The driver's identification of the material was inconsistent with placards and shipping papers. 'The truck was carrying diluted diphenylmethan disocyanate. Approximately 1,200 gal spilled. The trailer was not placarded and the shipping papers were destroyed in the wreck. The driver misidentified the contents of the load. After discussion with the carrier, the load was identified by responders. Fire fighters later determined that because of the low concentration of the material, placards were not required. The driver was cited for carrying improper shipping papers. Although the papers described the shipment as a mixed load containing flammable and corrosive solvents, they did not indicate which materials were flammable and which were corrosive. dGasoline tank truck ignited, destroying placards and shipping papers. Load consisted of wet storage batteries. The load was not placarded as required. 1Orange panel and UN/NA identification numbers were missing from truck. 8The truck was carrying 8,800 gal of molten sulfur, which was not regulated at that time. The truck did not have placards, although fire fighters treated the cargo as hazardous. "The truck was carrying a mixed load of combustible materials and Class C explosives. The vehicle was not placarded as required. The shipper was not aware that placards were required. 'The truck was carrying 50,000 lb of molten sulfur, which was not regulated at that time. The shipping papers could not be retrieved immediately because of the spill. The truck driver misidentified the material, causing fire fighters to evacuate 1,000 residents from the area. It took 3 hr for fire fighters to retrieve the shipping papers from the truck and properly identify the material. 'The load of petroleum-treating compound was placarded properly as a flammable liquid. The shipping papers described the load as "Flammable Liquid, n.o.s.," but no information was available on the component causing the hazard. Responders did not obtain this information for 4 hr. "The truck was carrying two 30-gal drums of herbicide. Because the material was not designated as hazardous by DOT, the truck was not placarded and shipping papers did not provide hazard information or the technical name of the material. Fire fighters were not aware that a potentially hazardous chemical was on board the vehicle until 1 hr after they began extinguishing the fire.
170 HAZARDOUS MATERIALS SHIPMENT INFORMATION were able to identify the load by contacting the carrier. Because of the low concentration of hazardous material, the truck did not require placards. A combination truck carrying a mixed load of corrosive and flammable solvents ran off the roadway and collided with a guardrail. The load of solvents was not damaged in the crash, and the driver received only minor injuries. The van trailer was placarded properly, but the shipping papers did not specify which of the solvents were corrosive and which were flammable. Because of the missing informa- tion, fire fighters were concerned about the accuracy of the shipping papers and their reliability in identifying the load. Fire officials had to contact the carrier and the shipper to determine the contents of each package in the vehicle. The driver of a combination tank truck containing 8,000 gal of gasoline lost control of the vehicle on an entrance ramp. The truck rolled over, killing the driver and causing the entire load to spill and ignite. The truck's placards and shipping papers were destroyed in the fire. Upon arrival at the incident, fire fighters were unsure of the contents of the vehicle, although they were soon able to make an identification because of markings that remained visible on the tank and the strong gasoline odor. 9. A truck carrying wet storage batteries collided with another truck, causing both vehicles to overturn and ignite. The truck contain- ing the batteries was not placarded as required. Fire fighters, who were extinguishing the fire, were not aware of the batteries until the shipping papers were retrieved from the cab, which was undamaged. The bat- teries were allowed to burn in the fire; no emergency response person- nel were harmed. 11. A combination flatbed truck carrying nearly 4,000 gal of com- bustible and flammable materials in eight portable tanks overturned after a sudden shift in the load. Several of the tanks were damaged, and nearly 400 gal spilled onto the roadway. There was no fire. Correct placards were displayed on the truck. However, the UN/NA identifica- tion numbers of the materials were missing, causing fire fighters to have to retrieve the shipping papers from the truck in order to fully identify the load. 14. The driver of a double-tank combination carrying 8,400 gal of molten sulfur momentarily lost control of the vehicle and struck a berm on the roadside. One of the tanks was damaged, causing a spill that resulted in a small fire and release of a vapor cloud. Responding fire
Data Needs and Sources 171 fighters could not immediately identify the material because the truck was not placarded. The shipping papers had to be retrieved from the cab to confidently identify the material. The truck was not placarded because molten sulfur was not designated as hazardous at the time. 15. A combination truck carrying a mixed load of flammable mate- rials and Class C explosives jackknifed and overturned. The accident resulted in extensive damage to the tractor and the trailer. Some of the packages containing the hazardous materials were ejected from the trailer, but they were not damaged and no material escaped. Because the truck was not placarded as required, fire fighters had to rely on the identification of the materials by the driver. 19. A tank truck damaged in an accident spilled 50,000 lb of molten sulfur onto the roadway. At the time molten sulfur was not regulated as a hazardous material; hence, the truck was not placarded. Because of the spill and ensuing fire, the truck's shipping papers could not be retrieved by responding fire fighters. The driver of the truck misiden- tified the load as sulfur tetrafluoride, which is a much more hazardous material than molten sulfur. This incorrect information caused local officials to close the highway and evacuate 1,000 people from the area. The actual cargo was not determined until 3 hr after the spill, when local fire officials contacted the carrier. 26. The driver of a tank truck carrying 5,000 gal of a compound used for refining crude oil lost control of the vehicle on a steep down- grade. The vehicle veered off the road and struck an embankment, killing a passenger in the truck and causing the entire load to spill. The truck was placarded correctly. However, the shipping papers did not disclose the chemical contents of the product, which was listed as "Flammable Liquid, n.o.s." Responding fire fighters needed more than 4 hr to determine the chemical name of the material and its hazards. 35. A flatbed straight truck carrying two 30-gal drums of herbicide in its mixed load veered off the road and rolled over after one of its tires blew out. The truck and its contents immediately caught fire. Respond- ing fire fighters fought the fire for about 1 hr before they were informed of the herbicide by the truck driver. The shipping papers, which had been reviewed by fire officials, contained only the brand name of the material and did not identify the material as herbicide or specify any hazards. Concerned that the herbicide might indeed be hazardous, fire fighters backed away from the fire and contacted the carrier and shipper for information. The shipper claimed that the truck did not require placards because the material was not designated as hazardous.
172 HAZARDOUS MATERIALS SHIPMENT INFORMATION NTSB Rail Cases In 1991 NTSB completed a major study of 45 hazardous materials rail incidents that occurred during the 12-month period March 1988 to February 1989 (NTSB 1991 a). The purpose of the investigations was to determine the factors that caused or contributed to the accidents and their outcomes and to recommend remedial actions for recurring prob- lems. NTSB documented the chronology of the responses, including the availability and effectiveness of placards, shipping papers, and other information sources. As with the NTSB truck study cited earlier, backup documentation for the cases was obtained by the committee and reviewed along with newspaper reports of the incidents. The documentation was sufficient to identify cases in which information problems occurred and the reported consequences of the problems. A summary of the cases is provided in Table B-3. The rail cases tend to have severe consequences, because many involve major crashes or derailments. In five cases, evidence indicates that emergency re- sponders encountered information problems. A summary of the five cases follows (cases are numbered according to their order in Ta- ble B-3): 2. A freight train had one of its hopper cars derail. The derailed car contained ammonium nitrate, which did not spill. A car carrying lumber separated the derailed car from two tank cars containing LPG. Because no cargo escaped, the railroad did not immediately notify local authorities. About 1 hr after the derailment, a witness alerted local fire officials. Upon arriving at the scene fire officials met the train crew, who identified the hazardous materials from waybill information and placards. Fire officials wanted to contact the shipper of the ammonium nitrate to verify the information and obtain a more detailed description of the product's hazards. However, the shipper's name and telephone number were not listed in the waybill, and the railroad did not have an emergency telephone number to provide assistance in obtaining this information. As a result, fire officials called CHEMTREC. CHEM- TREC communicators could not respond to the request immediately because fire officials did not know the shipper's identification number, consignee name, or hopper car number. Fire officials, who were not aware that this information was necessary, called CHEMTREC back with the name of the consignee. After 1 hr of making inquiries, CHEM-
TABLE B-3 Summary of 45 NTSB Hazardous Materials Rail Cases (NTSB 1991a) Hazardous Materials Cars Consequences Involved Hazard Class of Cargo No. of Information _____________________________________ Carrier Type Operat. Phase Problem Areas Tank/ Inter- FlamJ No. of People NTSB Case Class Hopper modal/ Comb./ Fatali- No. of Evacu- Firei SpiW Way- Case Number I Other Derail Other Total Car Flatcar Oxidizer Corr. Poison Other ties Injuries ated Expl. Vapor Placard bill Other 1 FFW-88-FRZ13 X X 4 X X 2 ATL-88-FRZ13 X X 3 X X 3 CHI-88-FRZI7 X X 4 X X 4 CHI-88-FRZI8 X X I X X 5 LAX-88-FRZ10 X X I X X 6 CHI-88-FRZ20 X X 3 X X X X 7 DEN-88-FRZI1 X X I X X 8 ATL-88-FRZ15 X X 6 X X X 9 CHI-88-FRZ22 X X I X X 10 LAX-88-FRZ12 X X I X X 11 ATL-88-FRZI6 X X 6 X X 12 Afl-88-FRZ19 X X 1 X X X 13 F-rW-88-FRZ23 X X I X X 14 DEN-88-FRZI7 X X 2 X X IS FTW-88-FRZ24 X X S X 16 DCA-88-MRZ06 X X 2 X X 17 FrW-88-FRZ25 X X 1 X X 18 FrW-88-FR226 X X S X X X 19 FrW-88-FR227 X X 13 X X 20 LAX-88-FRZ15 X X S X X X 21 FIW-88-FR228 X X 0 X X X 22 ATL-88-FRZ20 X X 7 X X X x 300 x x 2 X 1,000 X x 300 50 x x 12 4,000 X xx 1 X 4 X 2 2 1,500 X X 6 XX 30 X 4 70 X X 150 X X 600 2 300 X (continued on nt page)
TABLE B-3 (continued) Hazardous Materials Cars Consequences Involved Hazard Class of Cargo Information Carrier Type Operat. Tank/ Inter- Flam./ No. of N f o ° People Problem Areas NTSB Case Class Phase Hopper modal! CombJ Fatali- No. of Evacu- Fire! Spill Way- Case Number I Other Derail Other Total Car Flatcar Oxidizer Corr. Poison Other ties Injuries ated Expi. Vapor Placard bill Other 23 CHI-88-FR227 X. X 5 X X 300 24 ATL-88-FRZ2I X X 6 X X I X 25 ATL-88-FRZ22 X X 1 X X 4 10 X 26 AU-88-FRZ23 X X 4 X X X 27 CHI-88-FRZ29 X X S X X 30 X 28 Afl-89-FRZ02 X X 7 X X SO X 29 ATL-89-FRZ03 X X 10 X X X 130 X 30 CHI-89-FRZO5 X X I X X 31 FFW-89-FRZOI X X 6 X X X 300 X X 32 LAX-89-FRZ02 X X 1 X X 300 33 CHI-89-FRZ06 X X 2 X X 2 X 34 ATL-89-FRZ05 X X 3 X X X 1,000 X 35 CHI-89-FRZ07 X X 3 X X X X 20 X 36 FTW-89-FRZ04 X X 5 X X 200 X 37 CHI-89-FRZ68 X X I X X I X 38 NYC-89-FRZ03 X X I X X 10 X 39 LAX-89-FRZO5 X X 6 X X 500 X 40 LAX-89-FRZ13 X X I X X 1 16 500 X 41 DCA-89-MRZOI X X 3 X X 2 3,500 X X d d 42 CHI-89-FRZ1I X X I X X 4 X 43 LAX-89-FRZI5 X X 2 X X ISO
44 CHI-89-FRZ14 X X 10 X X 2 125 X 45 DCA-89-M2004 X - X - 9 X - X - - - - 55 1,785 X X Total 35 10 34 11 166 43 2 18 15 22 3 3 107 17,223 8 36 0 4 3 A hopper car containing ammonium nitrate derailed. Because no cargo escaped, the railroad did not notify local authorities. About 1 hr after the derailment, a witness alerted fire officials. Upon arriving at the scene, fire officials met the train crew, who identified the cargo from waybill and placard information. Fire officials needed to contact the shipper to verify the information and obtain a more detailed description of the product and its hazards. The shipper's name and telephone number were not listed in the waybill as required, and the railroad did not have an emergency telephone number to provide assistance in obtaining this information. Fire officials called CHEMTREC for assistance. CHEMTREC, however, could not respond to the request immediately because it needed the shipper's identification number, which was not available to fire officials. After 1 hr of making inquiries, CHEMTREC was able to contact the shipper, who provided the necessary information to local officials. 'A derailed tank car was breached, causing the release of 20,000 gal of petroleum sulfite waste. The tank car was not listed in the train consist as a hazardous materials car. Because placards were displayed, fire officials presumed correctly that the car contained hazardous materials. Because of missing consist and waybill information, they had to contact the railroad to obtain basic information about the identities of the materials and associated hazards. 'A tank car containing ethyl alcohol ruptured and released 15,000 gal. Because of incorrect and inconsistent information in the train's consist and waybills, the contents of the car were misidentified by fire fighters and railroad officials as propargyl. Because of remaining concerns about the accuracy of the identification, fire officials sought additional information before caking further action. Three hr after the derailment, additional evidence was found indicating that the car contained ethyl alcohol. The shipper was not identified and contacted to confirm the car's contents until more than 5 hr after the derailment. aA derailment and collision of three tank cars containing hydrogen peroxide and isopropyl alcohol resulted in a large explosion and fire. The railroad dispatcher at a nearby yard office attempted to notify local fire and police but could not get through because of an influx of calls by witnesses reporting the accident. Local officials did not contact the railroad. After nearly 1/2 hr, the railroad dispatcher was able to contact local police, who sent a patrol car to retrieve the documents, which were found to be inconsistent. Fire officials were given instructions by the railroad dispatcher to obtain product hazard information by contacting CHEMTREC, which was in communication with the shipper. CHEMTREC, however, did not receive a call from fire officials. 'Nine cars containing butane derailed and caught fire. After surveying the damage, one of the conductors removed the train's consist and waybills and left the accident site. Because of the smoke and fire, fire officials could not determine how many cars were involved in the accident and could not see the placards to identify the materials. The conductor, who was located by police, identified the hazardous materials as butane, but he lost the consist and waybills. After 30 mm, fire officials were able to contact railroad officials, who initially misidentified the cargo as butadiene. Because of the inconsistent identifications, fire officials requested copies of the consist and waybills. A second set of these documents had been available at the rear of the train; however, the rear cars had already been decoupled and moved to a rail yard 4mi from the accident. Local officials did not obtain the documents needed to confidently identify the materials until S hr after the derailment.
176 HAZARDOUS MATERIALS SHIPMENT INFORMATION TREC was able to contact the shipper. In the interim, fire officials evacuated 300 residents near the track as a precautionary measure. There was no fire or injuries. The evacuation was called off after the shipper was contacted and the specific product information was provided. iS. A freight train had 34 cars derail. A derailed tank car was breached and released its entire load of 20,000 gal of petroleum sulfite waste. The tank car displayed placards, alerting first responders to the presence of hazardous materials. In checking the train consist and waybills, however, local police did not find the car listed as containing hazardous materials. Because of the missing information, police offi- cials called the railroad to determine the exact contents of the derailed car and the associated hazards. The railroad provided the necessary information and the name and telephone number of the shipper with- out significant delay. The railroad notified CHEMTREC of the inci- dent, but further assistance by CHEMTREC was not needed. One family residing near the derailment was evacuated overnight. No one was injured in the incident. 31. A freight train had 22 cars derail. A derailed tank car containing ethyl alcohol ruptured and released 15,000 gal. Police officials arriving at the scene were not initially aware of the hazardous spill, but they detected a strong odor and immediately alerted fire officials to the possible involvement of dangerous chemicals. Several minutes later the railroad notified local police of the derailment and advised them of the possible involvement of dangerous chemicals. Because of the large number of cars involved, fire officials began evacuations before identi- fication of the spilled materials was complete. Approximately 300 people were evacuated. The train's crew provided responders with the consist and waybills and reported the positions of the derailed cars. The shipping documents indicated that the derailed tank car contained poisonous and flammable propargyl alcohol. Railroad officials, how- ever, were unable to identify and contact the shipper. Because of doubts about the accuracy of the shipping documents, fire officials sought additional information about the cargo before taking further action. Three hr later regional EPA officials uncovered information that indi- cated that the material was ethyl alcohol, which is less poisonous than propargyl. Five hr after the derailment, railroad officials were able to reach the shipper to verify the identification. 41. Forty-nine freight cars that had been decoupled rolled down a grade and struck another train. Fifteen cars from the runaway train
Data Needs and Sources 177 derailed, including tank cars containing hydrogen peroxide, isopropyl alcohol, and acetone. A fire and two explosions resulted. The train's crew retrieved the consist and waybills from the locomotive and ra- dioed a nearby yard office. The dispatcher at the yard office attempted to contact the local police, but the telephone lines were busy due to an influx of calls from witnesses reporting the accident. Local officials did not attempt to contact the railroad. After nearly 1/2 hr, the railroad dispatcher was able to reach local officials, inform them of the chemi- cals involved, and provide them with the consist and waybill docu- ments. The information in the consist and waybills, however, was inconsistent. Local officials were advised by the railroad to call CHEM- TREC to obtain specific product information and to communicate with the shippers of the various products. Local officials, however, did not attempt to contact CHEMTREC, resulting in a delay in obtaining important product information. Eventually, the Montana Department of Emergency Services contacted CHEMTREC for the information. (In its investigation of the accident, NTSB concluded that the railroad should have had a system in place for verifying the completeness and accuracy of shipping documents and that local officials would have benefited from specific instructions and procedures for responding to railroad accidents involving hazardous materials.) 45. A freight train had 21 of its cars derail, including 9 filled with butane. The nine cars came to rest next to a chemical plant. Two of the cars released butane, which ignited immediately. The crew at the front of the train notified the yard dispatcher of the accident and fire. After surveying the damage and consulting the train consist and waybills, the crew decoupled the locomotive and front cars and moved them forward 3/4 mi. In the process they lost the consist and waybills. In the meantime, fire officials had arrived at the site of derailment, but they could not determine how many cars were involved because of obstruction from the fire and the chemical plant. The train crew, which was located by a police officer, identified the contents of the derailed cars as butane. After 30 mm, fire officials were able to contact railroad officials, who initially misidentified the material as butadiene. Because of the incon- sistent identifications, fire officials requested copies of the shipping documents. A second consist and set of waybills had been available at the rear of the train; however, the rear cars had been decoupled by the rear train crew and moved 4 mi away to a rail yard. Five hr passed before fire officials received copies of the shipping documents.
178 HAZARDOUS MATERIALS SHIPMENT INFORMATION SUPPLEMENTAL INCIDENTS Although the 125 case studies provide a means of determining the frequency of information problems, the small number of cases cannot reveal all the possible kinds of problems. To supplement the cases, 12 additional hazardous materials incidents with information problems (8 truck and 4 rail) are described in Chapter 3. The incidents, which were derived from newspaper reports, were selected to illustrate types of information problems. Because they were not selected objectivelyâ that is, without prior knowledge of information problemsâthey are not used to determine the frequency of information problems. All involve injuries, evacuations, highway closings, environmental dam- age, or other extreme consequences; hence, they are not typical of most hazardous materials incidents. Supplemental Truck Incidents The following eight truck incidents were selected from more than 100 newspaper reports of hazardous materials truck accidents from 1988 to 1991. The newspaper reports (in some cases supplemented by NTSB reports) were obtained from a search of more than 40 newspapers across the county. In all cases missing or inaccurate emergency in- formation was identified as an important problem to emergency responders. Lorton, Virginia, July 23, 1991. A tractor-semitrailer carrying a mixed load of hazardous waste spontaneously ignited while on an Interstate highway. Fire fighters were not confident that the shipping papers accurately listed the contents of the load and were concerned about possible chemical reactions between the various materials. The uncertainty contributed to the 6-hr delay in removing the vehicle and reopening the highway (Washington Times, July 24, 1991). Rockville, Maryland, December 15, 1989. A tractor-trailer (van) carrying a mixed load of hazardous and nonhazardous materials over- turned on an Interstate highway. The truck driver provided fire fighters with the shipping papers. However, fire fighters were concerned about possible inaccuracies in the shipping papers because of the many ship- per documents involved. Also, fire fighters could not determine whether any of the containers had leaked, and they were concerned
Data Needs and Sources 179 about a potential chemical reaction. The uncertainty contributed to the 20 hr needed to clear the incident and the decision to evacuate more than 100 families (Washington Post, Dec. 16, 1989). Dallas, Texas, August 1, 1989. A tractor-trailer (van) carrying a mixed load of hazardous and nonhazardous materials overturned on a congested downtown freeway. The driver was killed in the accident. Listed on the shipping papers was a 650-lb shipment of hexane ti- tanium trichloride. The truck was not placarded because federal reg- ulations do not require placarding for shipments under 1,000 lb. Responding fire fighters, who were concerned that the chemical could be flammable and toxic, closed the highway and evacuated more than 2,000 workers from downtown office buildings. Fire officials could not find the material in the DOT Emergency Response Guidebook, because it is listed separately according to its constituentsâhexane trichloride and titanium trichloride. The presence of a poison placard in the wreckage, which was flipped accidently by the impact, contributed to the confusion (Dallas Times-Herald, Aug. 3, 1989). Beltsville, Maryland, March 3, 1988. A tractor-trailer carrying a mixed load of hazardous materials overturned on an Interstate high- way, resulting in the release of some material as vapor. Fire officials identified the contents of the vehicle from shipping papers and placards but were concerned about a possible chemical reaction, because they could not determine which containers were leaking. The uncertainty caused them to evacuate 450 homes and close the highway for more than S hr (Washington Post, March 28, 1988). Camp Pendleton, California, April 1, 1989. A tank truck contain- ing 4,000 gal of gasoline overturned and caught fire on an Interstate highway. The driver was killed in the collision and fire. Fire fighters responding to the accident could not determine the contents of the tank bccause of the flames surrounding it. No markings or placards were recognizable because of the smoke and fire. Because they could not determine the contents, fire officials closed the highway and the let the fire burn itself out rather than attempt to extinguish it. The highway was closed for 11 hr (Los Angeles Times, April 2, 1989). Salisbury, North Carolina, February 2, 1990. A tractor-trailer carrying low-level radioactive material was involved in a traffic acci- dent on an Interstate highway. The driver, who escaped before the truck caught fire, did not retrieve the shipping papers. When fire fighters arrived, only the frame of the truck remained recognizable. The driver immediately identified the material as radioactive but did not have
180 HAZARDOUS MATERIALS SHIPMENT INFORMATION additional information about the cargo. Assuming the worst, local authorities evacuated 100 people from the area and closed the highway. The trucking company was contacted but could not find the shipping papers. It took 5 hr for the fire fighters to make confident identification of the material, which was not harmful (Charlotte Observer, Feb. 10, 1990). Cleveland, Ohio, March 15, 1991. A truck carrying a mixture of nine corrosive, flammable, and poisonous materials caught fire on an Interstate in a densely populated area. The truck's shipping papers were incomplete and did not properly identify the packaging or hazard classes of all the hazardous materials being carried. Local officials contacted the carrier but were not confident that the information provided was accurate. The chemicals in the truck were contained in many different types of packages, and it was not possible to determine which containers were leaking and causing the fires and small explo- sions. Fire officials eventually contacted each of the nine shippers individually to determine the contents of the load. More than 2,000 people were evacuated from nearby residences, and the highway was closed for 36 hr during the event (Cleveland Plain Dealer, March 25, 1991). Collier County, Florida, November 30, 1988. A tractor-flatbed semitrailer carrying 32 cylinders of a poisonous and toxic-by- inhalation mixture overturned at the intersection of two farm roads in a sparsely populated area. Several of the cylinders were ejected and struck trees in the wooded areas adjacent to the accident site. State police officers, who arrived first at the scene, were not aware that the truck's cargo was hazardous. Although the vehicle was correctly placarded Poison, none of the placards was readily visible to the offi- cers, who approached the overturned vehicle to assist the injured driver. Only the placard on the front of the semitrailer was partially visible, but it was not identified until after one of the officers had climbed onto the vehicle to assist the driver. One of the cylinders was punctured, resulting in the release of toxic fumes near the vehicle. The police officers inhaled the fumes before realizing the danger. One of the officers experienced burning eyes, chest pains, and difficulty breathing. The officer and three other responders, who experienced similar symp- toms, were transported to the hospital for treatment. Shipping papers were not available to responders until after the incident because they were located inside the wrecked tractor (NTSB 1990).
Data Needs and Sources 181 Supplemental Rail Incidents The following supplemental rail incidents were selected from a review of newspaper reports and NTSB investigations of rail incidents during the past 5 years. Dunsmuir, California, July 14, 1991. A 97-car freight train had 6 of its cars and rear locomotives derail on a bridge over the Sacramento River. A derailed tank car was damaged and released 19,500 gal of the pesticide metam sodium into the river. Metam sodium is not designated by DOT as a hazardous material, and placards were not displayed on the tank car. Shipping papers for the car listed the material as a weed killer but did not specify potential hazards. In addition, the material is not listed in the DOT Emergency Response Guidebook. As a result, fire and police officials responding to the derailment were uncertain of the hazards associated with the material and did not have information indicating its toxicity to fish and other water life. Because of the uncertainty about the chemical, responders treated the spill as poten- tially hazardous and contacted CHEMTREC for safety data. More than 3 hr passed before responders were able to obtain basic informa- tion about the chemical's hazards (unpublished NTSB rail accident investigation documentation). Freeland, Michigan, July 22, 1989. A freight train had 14 of its cars derail. Six tank cars were damaged and released all or part of their loads and caught fire. The tank cars contained styrene, monomer, acrylonitrile, acrylic acid, naphtha, and a mixture of chiorosilanes. After the accident, the conductor radioed the rail yard dispatcher, who notified the local fire department. Fire officials arriving on the scene were immediately provided with the train consist and waybills. Ship- pers of the materials were notified, and many sent representatives to the scene within 2 hr. Railroad employees and a shipper representative assisted fire officials in reviewing the shipping documents and identify- ing the hazardous materials. The materials involved in the wreck were quickly identified. However, because the tank cars were jumbled in the pileup and placards and car markings were not visible due to smoke and fire, fire officials could not determine which materials were burn- ing. Without this information, fire fighters could not determine the appropriate extinguishing agent and were forced to let the fire burn longer than they otherwise might have. About 1,000 residents were evacuated almost immediately after the derailment (NTSB 1991b).
182 HAZARDOUS MATERIALS SHIPMENT INFORMA11ON Seacliff, California, July 28, 1991. Twelve cars of a freight train derailed under a freeway overpass. Many of the cars were flatbeds that had been carrying intermodal containers. In the pileup caused by the derailment, many of the intermodal containers were ejected. Fifteen 55-gal drums escaped from the containers and were punctured, spilling 440 gal of hydrazine. A 5,000-gal tank of naphthalene was also ejected near the drums, although it was not breached. Because fire units had to approach the train from the rear, they did not have access to the train's consist and waybills, which were located at the front end. The only access road leading to the front of the train passed through the accident site. This logistical problem resulted in a delay before fire officials could review the shipping documents, which were obtained by helicopter. After reviewing the shipping documents, fire officials were able to determine which materials were involved in the derailment. However, because many of the intermodal containers and their contents (drums) had been ejected from the cars and scattered, fire officials could not determine the origin and contents of each drum. To obtain this infor- mation, fire fighters in protective equipment had to visually inspect the labels on each of the damaged drums (unpublished NTSB rail accident investigation documentation). Miamisburg, Ohio, July 8, 1986. Fifteen cars of a 44-car freight train derailed. Three of the derailed cars were tank cars containing yellow phosphorus, molten sulfur, and tallow. The tank cars, which derailed on a bridge, were extensively damaged, lost product, and were involved in the resulting fire. Local fire and police officials were alerted by witnesses to the derailment. The first responders could only see two of the tank cars because of the smoke and fire. The more visible car had lost its placards in the crash; therefore, first responders were not immediately aware that hazardous materials were involved. After sur- veying the damage, the conductor met the emergency responders, alerted them to the presence of hazardous materials, and presented them with the train consist. The conductor had to return to the locomo- tive to retrieve the waybills, which he had inadvertently left behind. The waybills were not in order and had to be reassembled to identify all the cars in the derailment. In addition, the conductor left behind in the locomotive an emergency guide that accompanied the phosphorus shipment. To confidently identify all of the cars involved, fire fighters were sent into the derailment site in full protective equipment to visually inspect each car (NTSB 1987).
Data Needs and Sources 183 NOTE 1. The existing system is defined broadly to include the full array of emergency information sources and resources available to responders at the scene of an incident, such as placards, shipping papers, container labels and packages, the DOT Emergency Response Guidebook, the Material Safety Data Sheet, and CHEMTREC. REFERENCES NTSB. 1987. Hazardous Materials Release Following the Derailment of Baltimore and Ohio Railroad Company Train No. SLFR, Miamisburg, Ohio, July 8, 1986. Report NTSBIHZM-87/01. Washington, D.C. NTSB. 1988. Safety Study: Case Studies of 189 Heavy Truck Accident Investiga- tions. Report NTSB/SS-88-05. Washington, D.C. NTSB. 1990. Puncture of a Cylinder Containing a Mixture of Methyl Bromide and Chioropicrin Following the Overturn of a Tractor/Semitrailer, Collier County, Florida, November 30, 1988. Report NTSBIHZM-90/01. Washington, D.C. NTSB. 1991a. Safety Study: Transport of Hazardous Materials by Rail. Report NTSB/SS-91-01. Washington, D.C. NTSB. 1991b. Derailment of CSX Transportation Inc. Freight Train and Haz- ardous Materials Release Near Freeland, Michigan on July 22, 1989. Washing- ton, D.C.
APPENDIX C National Transportation Safety Board Recommendations Pertaining to Emergency Response Information (1980 to 1992) The National Transportation Safety Board (NTSB) has provided the following list of recommendations made since 1980 concerning emergency information, planning, and training as a result of its haz- ardous materials incident investigations and safety studies. SHIPPING PAPERS R-90-03 1 (rail): NTSB recommends that CSX Transportation, Inc., require train crews to update the train consist to accurately reflect the position of hazardous materials as cars are added or set off en route. R-88-067 (rail): NTSB recommends that Mitsui and Company (USA) Inc. enter on its shipping papers for hazardous materials a 24-hr telephone number where detailed knowledge of the hazardous charac- teristics of the materials being shipped can be obtained. R-87-018 (rail): NTSB recommends that the Research and Special Programs Administration require that on cabooseless trains involving hazardous materials a conspicuous weatherproof container be affixed at, on, or near the rear-end marker to hold a current consist list for use by emergency response forces. 1-86-004 (intermodal): NTSB recommends that the Department of Defense identify the relative amounts of hazardous ingredients con- 184
National Transportation Safety Board Recommendations 185 tamed in Department of Defense waste shipments and provide that information with the shipping papers to better inform emergency re- sponse personnel about the composition and hazards of the waste material being transported; include action that can be taken to mitigate the shipments' hazards. H-85-044 (highway): NTSB recommends that RISS International Corporation enter the RISS 24-hr emergency telephone numbers on shipping documents prepared for Class A and B explosive and other high-hazard shipments and instruct drivers to provide the telephone numbers to emergency response personnel in case of an accident. 1-85-023(intermodal): NTSB recommends that the U.S. Depart- ment of Defense amend the "special instructions for motor vehicle drivers" that accompany explosive and other high-hazard Department of Defense shipments to provide local emergency response personnel comprehensive shipment-specific hazardous information and ship- ment-specific preautionary actions, including appropriate evacuation distances. 1-85-010 (intermodal): NTSB recommends that the Research and Special Programs Administration determine the adequacy of general shipping names on shipping papers for hazardous wastes and the need for additional information, such as technical and chemical group names, to better inform emergency response personnel about the com- position and hazard of the material being shipped. 1-85-004 (intermodal): NTSB recommends that the Harris Corpora- tion enter information on shipping papers to better inform emergency response personnel about the composition and hazards of the waste material being shipped, as permitted by Title 49 CFR 172.202, and include action that can be taken to neutralize the material and mitigate its hazards. R-83-005 (rail): NTSB recommends that the Research and Special Programs Administration require that all shipping papers accompany- ing hazardous materials in trailer-on-flatcar or container-on-flatcar shipments identify the originating shipper of the hazardous materials in order to facilitate access to technical emergency response information in the event of a hazardous materials incident. 1-83-002 (intermodal): NTSB recommends that the U.S. Depart- ment of Transportation, Research and Special Programs Administra- tion, determine, by mode of transportation, the feasibility of requiring comprehensive product-specific emergency response information, such as materials safety data sheets, to be appended to shipping documents
186 HAZARDOUS MATERIALS SHIPMENT INFORMATION for hazardous materials transported in bulk quantities, giving particu- lar attention to the early emergency response problems posed by N.O.S. commodities in transit. For those modes of transportation for which a positive determination results, incorporate necessary require- ments into Title 49 of the code of federal regulations. R-83-002 (rail): NTSB recommends that the Southern Pacific Trans- portation Company improve current total operations processing sys- tem procedures to better assure that train crews of trains carrying hazardous materials are furnished accurate information regarding the train consist and the appropriate emergency response for the haz- ardous material. EMERGENCY PLANNING R-91-015 (rail): NTSB recommends that major railroads develop, implement, and keep current, in coordination with communities adja- cent to railroad yards and along hazardous materials routes, written emergency response plans and procedures for handling release of haz- ardous materials. The procedures should address, at a minimum, key railroad personnel and means of contact, procedures to identify the hazardous materials being transported, identification of resources for technical assistance that may be needed during the response effort, procedures for coordination of activities between railroad and emer- gency response personnel, and the conduct of disaster drills or other appropriate methods to test emergency response plans. R-8 9-029 (rail): NTSB recommends that the American Short Line Railroad Association encourage all member railroads to maintain a 24- hr telephone number and a point of contact for public and emergency service use in the event of an emergency or dangerous situation. R-88-069 and R-88-070 (rail): NTSB recommends that the National League of Cities and the National Governors' Association urge their members in coordination with rail yard management to develop and implement emergency response procedures for handling releases of hazardous materials from railroad vehicles. R8 8-OS 7 (rail): NTSB recommends that Norfolk Southern imple- ment a procedure for periodically determining that its rail yard man- agement maintains up-to-date emergency notification listings for local emergency response agencies and that employees required to have these lists have current information and know when and how to use them.
National Transportation Safety Board Recommendations 187 R-85-098 (rail): NTSB recommends that the Arkansas Office of Emergency Services develop in cooperation with the railroads oper- ating in the state and with the Federal Emergency Management Agency emergency response guidelines for use by communities adjacent to railroad yards that handle bulk shipments of hazardous materials and assist those communities in the development of effective procedures for responding to releases of hazardous materials within railroad yards. The procedures should address, at a minimum, initial notification, identification of key contact personnel, response actions for the safe handling of releases of the various types of hazardous materials trans- ported, identification of the resources to be provided, actions to be taken by the railroad and by the community, and emergency drills and exercises. R-85-054 (rail): NTSB recommends that the Federal Railroad Admin- istration develop in cooperation with the railroad industry and the Federal Emergency Management Agency criteria for emergency planning and response for use by operators of railroad yards that handle bulk shipment of hazardous materials and incorporate these criteria into applicable sections of 49 CFR. The criteria should address, at a minimum, emer- gency plan content, initial notification procedures, the conduct of emer- gency drills and exercises, and the coordination of planning and response activities with local public emergency response officials. R-85-050 (rail): NTSB recommends that the Missouri Pacific Rail- road Company, using its master railroad yard emergency response guideline and in coordination with communities adjacent to its railroad yards, develop local emergency and response plans appropriate to the hazards attending the conditions and operations within each of its railroad yards. R-80-007 (rail): NTSB recommends that the Federal Railroad Ad- ministration require operating railroads to develop emergency re- sponse plans, put them into effect, and file those plans with the Federal Railroad Administration in a similar manner as is required by 49 CFR 217 with respect to operating rules. OPERATOR (TRAIN CREW/TRUCK DRIVER) ACTIONS/TRAINING R-90-039 (rail): NTSB recommends that the Federal Railroad Admin- istration require that Federal Railroad Administration personnel re- sponding to a derailment involving hazardous materials (a) make their
188 HAZARDOUS MATERIALS SHIPMENT INFORMATION presence and purpose known to local emergency response personnel, (b) advise local authorities of guidance available for assessing tank car damage and wreckage clearing operations, and (c) notify emergency response personnel of any imminently hazardous conditions that may exist. R-90-028 (rail): NTSB recommends that CSX Transportation, Inc., provide training, in addition to operating rules classes, to operating crews and supervisors on the actions they are to take immediately following an accident involving hazardous materials. This training should include, at a minimum, (a) the responsibility of crew members to identify themselves to emergency response personnel and to provide accurate information, including on-board documentation, of haz- ardous materials involved in the accident, (b) the responsibility of supervisory personnel to verify that emergency response personnel have all needed information and that it is accurate, and (c) the means by which supervisors are to determine if employees understand fully their responsibilities. R-87-056 (rail): NTSB recommends that CSX Transportation reem- phasize to all operating personnel the importance of directing their initial activities following a derailment to the cooperative support of local emergency response agencies. R-8 7-OS 5 (rail): NTSB recommends that CSX Transportation estab- lish crew member procedures for providing timely, accurate informa- tion to on-scene emergency responders about the types of hazardous materials being transported in tank cars involved in derailments where "end-of-train" devices are used. 1-85-006 (intermodal): NTSB recommends that Chemical Waste Management, Inc., evaluate company training and instruction pro- cedures and revise them, as necessary, to instruct drivers in the recog- nition of problems that may be incurred during transportation of hazardous wastes and in the notification of emergency response personnel. R-85-079 (rail): NTSB recommends that the Seaboard System Rail- road require its railroad dispatchers to notify local emergency response agencies immediately of a derailment of a train transporting hazardous materials. R-83-049 (rail): NTSB recommends that the Seaboard System Rail- road require supervisory personnel arriving at the scene of an emer- gency to determine what information has been provided by train crews to emergency response personnel, to verify the accuracy of the informa-
National Transportation Safety Board Recommendations 189 tion provided, and to advise the on-scene coordinator of any errors or omissions in the initial information given by the train crew. R-83-048 (rail): NTSB recommends that the Seaboard System Rail- road periodically instruct and test train crews and supervisory person- nel on the procedures for using train documents to identify all cars transporting hazardous materials and the information to be provided to assist emergency response personnel. R-83-008 and R-83-011 (rail): NTSB recommends that the Associa- tion of American Railroads and the American Short Line Railroad Association recommend to their member railroads that they assess their procedures to make certain that train crews of trains carrying hazardous materials have in their possession accurate documentation of, and emergency response information for, all hazardous materials being carried. 1-83-003 (intermodal): NTSB recommends that Matlack, Inc., revise its driver's manual to include information on the purpose and intent of making shipping papers continuously available to response personnel at the scene of a hazardous materials accident and thoroughly instruct its employees regarding these procedures. R-80-006 (rail): NTSB recommends that the Federal Railroad Ad- ministration develop and validate through simulated disaster exercises a model emergency response plan for the guidance of the railroad industry in formulating individual plans to be utilized by their train crew members in the event of emergency. RESPONDER ACTIONS/TRAINING R-90-040 (rail): NTSB recommends that the International Association of Fire Chiefs urge its members to emphasize to the appropriate emer- gency response personnel the need (a) to locate all crew members and train documents as a priority action when responding to a train acci- dent and (b) to search out information about the severity of known tank car damage and dangers posed, possible solutions or alternatives, and risks involved. R-88-055 (rail): NTSB recommends that the city of New Orleans in coordination with local railroads, review and revise emergency re- sponse procedures to make them applicable for handling releases of hazardous materials from railroad vehicles. At a minimum, these pro- cedures should address initial notification procedures, response actions
190 HAZARDOUS MATERIALS SHIPMENT INFORMA11ON for the safe handling of releases of the various types of hazardous materials transported, identification of key contact personnel, conduct of emergency drills and exercises, and identification of the resources to be provided and of actions to be taken by the rail yard operators and the community. 1-83-001 (intermodal): NTSB recommends that the International Association of Fire Chiefs, the International Association of Chiefs of Police, and the International Society of Fire Service Instructors inform their memberships, through emphasis in training, publications, and other means, of the safety and command benefits in restricting the access of emergency response groups to hazardous materials accident sites until their potential exposure to safety hazards and the need for their response activities can be determined. OTHER 1-85-022 (intermodal): NTSB recommends that the U.S. Department of Defense amend emergency response procedures to provide to local emergency response personnel prompt, coordinated, on-scene emer- gency response assistance in accidents involving the transportation of explosives and other high-hazard Department of Defense shipments. 1-85-021 (intermodal): NTSB recommends that the U.S. Depart- ment of Defense establish an effective 24-hr communication system to provide local emergency response personnel immediate access to au- thoritative information and expertise on the threats presented by explo- sive and other high-hazard Department of Defense shipments involved in transportation accidents. 1-85-019 (intermodal): NTSB recommends that the Research and Special Programs Administration regulate molten sulfur and, as appro- priate, other molten materials as hazardous materials, prescribe pack- aging and handling standards, and incorporate information relating to the hazards of these materials into warning devices and publications available to emergency responders and others involved in the transpor- tation of molten materials.
APPENDIX [a] Information Technologies in Transportation Computer-based information systems in transportation perform functions similar to some that would be required of a hazardous materials information system. Understanding existing systems is criti- cal for assessing the feasibility of any proposed hazardous materials system, since new requirements would be less expensive, more practi- cal, and more reliable if they built on existing capabilities. Four related categories of systems in transportation settings may be relevant: electronic data interchange (EDI); shipment-tracking sys- tems; automatic electronic identification (AEI) of vehicles, containers, or packages; and geographic positioning systems. They are described in the following sections. EDI EDI is the digital electronic exchange of business documents using standard codes and formats, replacing paper transactions. In transpor- tation, for example, shipping instructions may be transmitted from the shipper to the carrier, the freight bill from carrier to shipper, and the payment back from shipper to carrier electronically. In addition to shipper-carrier transactions, the same EDI system may be used by each party internally and between two carriers to transmit documents for an interline or intermodal shipment (Prince 1990). 191
192 HAZARDOUS MATERIALS SHIPMENT INFORMATION ED! is used today by all major railroads and large shippers. All the larger trucking companies also use ED!, but it has not yet been exten- sively adopted among the tens of thousands of small trucking com- panies and the millions of small trucking customers in the country. A 1991 American Trucking Associations survey found 176 trucking com- panies using ED!, up from 86 in 1988. One of the largest, Yellow Freight, reported processing 12 percent of its freight bills electronically (Bell 1991). Standardization has been the key to the growth of ED!. Two U.S. organizations have been involved in transportation ED! standards, the American National Standards !nstitute (ANS!) and the Electronic Data !nterchange Association. The two groups coordinate activities, and the trend is toward ANSI taking on primary responsibilities. Domestic ED! systems increasingly conform to industrywide standards, but some nonuniformity still exists. !nternationally, a standard developed by the United Nations, ED!FACT, is gaining general acceptance (Prince 1990). EDT is significant for the hazardous materials information system study, because these systems contain on-line, real-time, standardized information about the contents of freight shipments. Some ED! appli- cations that might be models for hazardous materials applications include the following: The U.S. Customs Service has a congressionally mandated process under way to adopt ED! as the standard method of communication with shippers and carriers, automating the transmittal of manifests, invoices, bonds, and other documents. This is an example of govern- ment using EDT to streamline a regulatory procedure (McNamara 1991, 14). The Environmental Protection Agency (EPA) and the National Governors' Association (NGA) have a pilot project under way in Vir- ginia and Pennsylvania in which shippers of hazardous wastes transmit to the state via ED! all manifests for shipments of hazardous wastes. Many states have requirements for filing of hazardous waste manifests, but there is no requirement that the states' data be kept up to date on shipments while they are in progress. Nonetheless, the NGAIEPA pilot shows that an infrastructure is beginning to be installed that might be a basis for other applications (NGA 1991). Transport Canada and the Canadian railroads have tested a pro- cedure for giving local government emergency response officials direct
Information Technologies in Transportation 193 access to railroads' ED! systems for obtaining cargo information at the scene of rail accidents (McBrearty et al. 1991). SHIPMENT TRACKING Many transportation companies have systems that maintain current information about the location of each shipment in transit. The systems are used internally to control operations and externally as a service whereby a customer can learn at any time the whereabouts and ex- pected arrival time of its shipments. The customer's access to this service may be via ED!. Related systems keep track of the location of railcars, intermodal containers, and truck trailers. The systems consist of standard formats for identifying and describ- ing shipments, a series of checkpoints (e.g., terminals, breakbulks, or sorting facilities) where the presence of each shipment or container is reported as it passes through, telecommunications links, and a central computer repository for the data. Data entry may be manual or may use bar coding or radio frequency transponders. The identifier attached to the shipment serves as a link to other data bases containing shipping documents and information about the contents. Small package carriers such as Federal Express and large less-than- truckload trucking companies (which handle many small shipments) have been leaders in shipment tracking. The railroads can track most railcars. Railcar tracking is centralized in the Association of American Railroads' Train !I service, which allows subscribing railroads or ship- pers to learn the location of any car nationwide. An interindustry task force sponsored by the chemical and railroad industries is completing work on a plan to standardize methods of identifying and tracking all rail shipments of chemicals. Industry has suggested making the information base that will emerge through this activity available to emergency responders through CHEMTREC, so that, given a car number, train number, or location, CHEMTREC could immediately determine all materials on the train (Subcommittee on Surface Transportation 1990, 1062). AEI AEI can be used for a variety of shipment identification and tracking applications. A small, inexpensive, short-range, programmable trans-
194 HAZARDOUS MATERIALS SHIPMENT INFORMATION ponder attached to a container or vehicle emits a short digital message when stimulated by an external signal. The device may be battery powered or nonpowered (transmitting by absorbing power from the reading signal). An International Standards Organization standard for container transponders has recently been promulgated (HELP News 1991), and it is expected that within a few years nearly all intermodal containers and railcars will be equipped with transponders (Progres- sive Railroading 1991). Truck applications have so far been mainly experimental. Some applications of AEI are similar to bar coding, but the technol- ogyis more suitable for remote, automatic reading. In a typical applica- tion, transponders on containers in a marine container terminal can be automatically read each time the containers are handled as they move from flatcars to holding areas to ship decks. Transponders on cars and trucks are also in use for automatic highway toll debiting. In a hazardous materials information system, AEI might have appli- cation as a means for remote identification of vehicles (an "electronic license plate"), or the devices could be encoded with an identifier for the specific materials on board. A test of one such application is planned as part of Advantage 1-75, a multistate project demonstrating uses of AEI to streamline enforcement of interstate trucking regulations. In the test, participating trucks operating on Interstate 75 in the eastern United States will be equipped with transponders. Once a transponder- equipped truck entering the corridor had demonstrated at one inspec- tion station that it complied with weight and other regulations, it would be automatically identified and allowed to bypass inspection stations further along the route (Inside IVHS 1991a). In the proposed haz- ardous materials application of this system, the record maintained on each truck in the corridor would include identification of hazardous materials in its cargo. Alternatively, the transponders themselves would transmit a hazardous materials identification code (Costantino 1991). GEOGRAPHIC POSITIONING SYSTEMS Commercially available systems allow central dispatchers for nation- wide trucking companies to continuously monitor the location of all their trucks, using satellite navigation and communication and radio transmitters on the trucks. Market penetration of these systems is
Information Technologies in Transportation 195 growing, but only a tiny fraction of the intercity truck fleet is equipped at present (Inside IVHS 1991b). In addition to the commercial systems, Sandia National Laboratory has developed the Security Tracking and Response Base (STARBASE) for the Department of Energy. This is a two-way geographic position- ing system for real-time tracking of shipments of selected radioactive materials. The purpose of the system is to ensure reliable, rapid emer- gency response in the event of an incident involving the transport of these materials. To that end, a system has been developed that takes advantage of commercial geographic positioning services and aug- ments them with other navigation systems, geographic information system displays, decision aids, mass storage systems, and backup com- munications. The Department of Defense has a similar system for tracking the transport of certain munitions, and the Department of Energy is developing a system for tracking the transport of radioactive wastes to proposed storage sites in New Mexico. The Drug Enforcement Administration, the Customs Service, and other federal agencies are investigating the use of such vehicle tracking devices for law enforcement applications (Inside IVHS 1991c). A system for monitoring the geographic position of hazardous mate- rials shipments would be more elaborate and probably much more expensive than the original proposal for the central reporting system, which would contain no information other than the fact that a ship- ment was in transit and perhaps a projected route. REFERENCES ABBREVIA11ON NGA National Governors' Association Bell, L. A. 1991. Carriers Show off EDI Expertise. Traffic World, May 27. Costantino, J. 1991. IVHS Progress Report. MVMA Government-Industry Re- search Symposium, July 10. HELP News. 1991. New Container Freight Tags. Fall. Inside IVHS. 1991a. FHWA Considers Truck Transponder Project for 1-80. May 28. Inside IVHS. 1991b. Qualcomm Nets Half of Geostar Users: 2,000 Units Convert. May 28. Inside IVHS. 1991c. U.S. Government Sets Sights on Tracking Systems. May 28.
196 HAZARDOUS MATERIALS SHIPMENT INFORMATION McBrearty, K., M. Mel!, and A. Carny. 1991. Electronic Waybill for Rail Transport of Dangerous Goods. Publication TP1 11064. Transport Canada, Ottawa, On- tario, Canada, Aug. McNamara, J. 1991. U.S. Customs Service Is Counting on EDI. Transport Topics, June 17. NGA. 1991. Backgrounder: Electronic Reporting of Biennial Report Data: NGA Seeking States for Pilots. Sept. 23. Prince, T. 1990. EDI Poses Promises, Problems. Progressive Railroading, June. Progressive Railroading. 1991. AAR Order Gives AEI Momentum. Dec. Subcommittee on Surface Transportation. 1990. Hearings. Transportation of Haz- ardous Materials (101-25), Hearings Before the Subcommittee on Surface Trans- portation of the Committee on Public Works and Transportation, House of Representatives, USGPO.
APPENDIX E Shipper, Carrier, and Emergency Responder Information Capabilities I f new requirements to maintain or report hazardous materials ship-ment information are enacted, shippers and carriers will need inter- nal information systems to ensure that the requirements are met. Creating and operating the systems would be a major cost of the new requirements. The feasibility of implementing the systems on a wide- spread basis throughout the shipper and carrier industries would deter- mine the practicality of the requirements and the time necessary for implementation. In addition, emergency responders will need information system capabilities to be able to use any new automated information service. The final section of this appendix characterizes the capabilities of first responders, hazardous materials emergency teams, and emergency responder dispatch centers. SHIPPER AND CARRIER CASE STUDIES To investigate what measures would be needed to comply with new information reporting requirements, interviews were conducted with selected shippers and carriers. They are presented as case studies. The objective of the case studies was to determine the following: 197
198 HAZARDOUS MATERIALS SHIPMENT INFORMATION Shipment-tracking capabilitiesâthe company was asked to de- scribe its ability to provide information about the contents of vehicles holding its shipments; Uses of electronic data interchange (EDI)âthe respondent de- scribed current applications of ED! in transportation or shipping, the fraction of transactions conducted via ED!, and use of EDT for bills of lading or shipment manifests; Impacts of new information requirementsâthe respondent de- scribed the actions that would be necessary if the company were obliged to participate in a central reporting system or other new arrangement. Answers to the questions about current shipment-tracking capa- bilities and uses of ED! help determine the kinds of costs a carrier or shipper would incur in participating in an automated hazardous mate- rials information system. If the company already keeps a real-time record of which shipments are in which vehicles, it can make that information available by reporting it to a central data base or in response to an inquiry from a central clearinghouse. Similarly, if the company uses EDT to transmit similar shipment information or has the facility for automated response to customer inquiries about shipment status, the communications capabilities for full participation in an automated hazardous materials information system may be largely in place. On the other hand, if a company does not have the relevant shipment-tracking or ED! capabilities, they would have to be specially developed to comply with new requirements. The case study shippers and carriers are described in the following sections. The firms include representatives of the following sectors: chemi- cal distributor, small package carrier, railroad, less-than-truckload (Lii) carrier, chemical manufacturer, and air freight carrier. Most of the firms described are typical of their segments of the hazardous materials han- dling industries and indicate the kinds of problems and costs that imple- menting a computer-based hazardous materials shipment information system would entail. However, shipper and carrier procedures and capa- bilities can vary substantially even within industry segments. The case studies support the following conclusions regarding feasibility and cost: If feasibility is defined to mean that firms would be technically capable of developing the necessary information systems at a cost not
Shipper, Carrier, and Responder Information Capabilities 199 deeply disruptive of the industry, the central reporting system or the clearinghouse option is feasible today in some segments of the carrier and shipper industries. For example, compliance would be feasible for most rail freight because of widespread implementation of car-tracking capabilities. However, in other sectors (for example, small package carriers) many firms have little or no shipment-tracking capability today, and imposition of a central reporting system requirement would force them to greatly alter their freight-handling systems or refuse hazardous materials shipments. Throughout freight transportation, carriers are experiencing great customer demand for shipment-tracking information to support just- in-time production and reduce inventory costs. Carriers and shippers are rapidly developing new capabilities for keeping account of goods in transit and efficiently using vehicle fleets. The feasibility of a technology-based information system is en- hanced by the existence of internal shipper and carrier systems. The capabilities of systems in use overlap significantly with those that might be required for an information system to support emergency response. In addition, the trend is for these systems to proliferate, to become more capable, and to become cheaper. All shippers and carriers would have to develop new procedures to keep continuous records of the vehicles in which shipments are con- tained, and most would have to expand ED! capabilities, before they could comply with the requirements of a central reporting system or a clearinghouse system. The cost of a hazardous materials transportation information system will depend on the functions it supports, the level of automation introduced, the ability to take advantage of existing and planned systems, and the degree of shipper and carrier participation required. Of the major components of costsâshipper and carrier internal infor- mation systems, the central service facility, the emergency responder interface, and enforcement and trainingâthe cost of shipper and car- rier internal systems could be the largest and would certainly be greater than the cost of the central facility. In many circumstances, the carrier is in a better position than the shipper to identify the contents of a vehicle. For example, a package, LTL, or intermodal shipment normally is carried by several vehicles while in transit, and only carriers can track vehicle changes for a single shipment. On the railroads, car-tracking capabilities are maintained by
200 HAZARDOUS MATERIALS SHIPMENT INFORMATION the carriers. In truckload and intermodal carriage of time-sensitive freight, carriers are developing tracking capabilities as a service to their customers. Capabilities differ greatly by size of firm, commodity, and mode. Within a sector, new information reporting requirements would disad- vantage smaller firms, because development of the internal systems needed to comply with the requirement would be in part a fixed cost and therefore a smaller percentage cost increase for large firms. How- ever, the interviews suggest that even small firms in some sectors would be in a better position to comply today than the largest firms in other sectors. Institution of a central reporting system or clearinghouse for rail shipments would be the most straightforward phase of implementation because of rail's existing car-tracking system, although there are gaps even in this system. CHEMICAL DISTRIBUTOR Company Description Chemical distributors may be taken as representative of a broad range of small business shippers and carriers of hazardous materials. Distrib- utors buy chemicals from primary producers and sell them to smaller customers, often after repackaging. They may blend materials to cus- tomers' requirements, but in general they are not manufacturers. Most distributors operate their own fleets of tractor-semitrailers for cus- tomer delivery.. The industry includes hundreds of small firms serving regional and local markets and a few large firms with nationwide systems of distri- bution centers. The companies interviewed are roughly the median size in the industry. Two distributors were interviewed. The one described here has an- nual sales of $45 million, SS employees, and two shipping locations. It mainly serves customers within a 200-mi radius in the Northeast. It sells about 1,000 products, of which about 75 account for most busi- ness. Among its principal products are solvents and other materials used in producing paints and coatings. The company operates a fleet of 6 tractors and 12 semitrailers, including tankers and vans.
Shipper, Carrier, and Responder Information Capabilities 201 Most liquid materials are delivered to the company from manufac- turers by tank truck. The company also receives some rail tank carload shipments. All inbound shipments of dry materials are in packaged form via van trailer. Inbound dry van shipments are loaded at pro- ducers' docks and unloaded at the distributor's warehouse without any intermediate handling of the freight. The typical inbound truck arrives with a mixed load of many packages and materials. All outbound shipments to the company's customers are by truck, either bulk liquids in tank trailers or packaged dry materials and liquids in drums and cans aboard van trailers. Most tank trailers are compart- mented, so deliveries of up to six bulk liquid products can be made on one run. Nearly all liquid bulk deliveries and most dry material deliv- eries are made by the company's fleet of tractors and trailers. Remain- ing deliveries are made by for-hire motor carriers. A company tractor-semitrailer carrying dry chemicals may deliver an entire truckload to a single customer, but more typically it stops at six to eight customers during a 1-day run. It may carry dozens to hundreds of individual packages and many different materials. The median shipment size is about 1 ton but varies greatly among the company's various lines of business. Company trucks are often able to backhaul a shipment from a chemical producer to the distributor. Roughly 80 percent of the company's departing trucks carry at least one hazardous material regulated by the U.S. Department of Transpor- tation (DOT) on departure, although many of the chemicals the com- pany sells are not regulated by DOT. The most common hazard class is flammable liquid. Steps in Handling Orders and Shipments The distributor receives an order by phone or fax from a customer. The customer service clerk enters the order in the company's computerized order entry system. The company does not use ED! to receive orders from customers or to place orders with suppliers. A bill of lading is generated by the computer and transmitted to a terminal at the appropriate shipping location. If any of the materials ordered are DOT-regulated hazardous materials, the computer also generates the legally required identifying information on the bill of
202 HAZARDOUS MAThMAt.S SHIPMENT INFORMATION lading. The company must regularly update its hazardous materials data base to ensure that the computer-generated bills reflect current regulations. 4. At the warehouse, the dispatcher sorts bills of lading into trailer loads, considering quantity of material and destination. S. Trailers are loaded each day according to the bills the dispatcher has assigned to each trailer. The driver is then given the bills of lading for his load. Hazardous materials bills have a tab affixed to their edges so they may be readily pulled from the pile. The driver also receives a list of each delivery stop on the day's run in order, and a list of the bill of lading numbers for each stop. The driver placards the trailer according to the information on hazard class and quantity printed on the bills of lading. The driver then makes the day's deliveries. Drivers are instructed to call the dispatcher at set times, usually midafternoon and at the completion of all deliveries. After the last delivery, the driver picks up the back haul, if there is one, and returns to the distributor, usually the same day. The driver returns with a copy of each bill of lading for a com- pleted delivery signed by the receiver. These are entered to update the order entry system. Any differences between the materials and quan- tities originally ordered and what was actually delivered are recorded. Entry of the record of the completed shipment triggers billing, and the record is then dropped from the order entry system. Requirements for Participating in a Central Reporting System or Clearinghouse The following actions are one approach to developing the information system and communication capability that the company would need to participate: The functions of the order entry data base would have to be expanded to keep account of which trailer each shipment in transit was aboard. As cach trailer was loaded, employees at the loading area would have to make an entry at a computer terminal updating the record for each order (corresponding to the bill of lading for that order)
Shipper, Carrier, and Responder Information Capabilities 203 to show an identification number for the trailer holding the shipment. There is already a terminal at the loading dock that is used to receive bills of lading for the day's shipments and to plan dispatches for the day (that is, a file contains a list of bill of lading numbers assigned to each driver departing that day). Software would be needed to generate manifests showing the contents of each trailer at any time. Each driver would have to call the dispatcher after each delivery to report that a shipment had been completed and verify that the materials and quantities shown on the bill of lading were those actually delivered. A central office clerk would then immediately update the order entry data base to show that the shipment was no longer on the trailer. Alternatively, if a central reporting system were in place, the burden of recording the delivery of a shipment could be on the receiver. The company would have, to install the ED! capability needed to transmit records to the central reporting system or, in the case of a clearinghouse system, to receive inquiries to its order data base from the clearinghouse. An update to a central reporting system data base would have to be transmitted each time a truck departed the warehouse and each time it made a delivery. The company also frequently uses for-hire carriers, both truckload and LTL, for deliveries to its customers. It could record an identifier for the carrier truck that picked up these shipments but would have no practical way of recording transfers of shipments to other vehicles. Cost of Participating in a Central Reporting System or Clearinghouse The principal cost elements to the company would be Development costs for modifying the order entry system to keep track of shipment loading, Central office labor cost for entering shipment loading and un- loading transactions, Driver labor for reporting shipment deliveries, Communications cost for reporting deliveries, Installation cost for ED! hardware and software, and ED! communications cost.
204 HAZARDOUS MATERIALS SHIPMENT INFORMATION SMALL PACKAGE CARRIER Industry Description Ground transportation of small packages is a specialized segment of the LTL freight market. These carriers handle high volumes of small pack- ages, with a maximum size limit (usually 70 lb) imposed. They operate hub-and-spoke systems: packages are picked up in smaller vehicles and carried to local terminals, where they are loaded onto tractor- semitrailers for carriage to a larger hub terminal. A typical shipment might be in five trucks or trailers and pass through four terminals. The one or two dominant private firms in the industry compete with smaller private firms and parcel post and to some extent with air express carriers and conventional LTL. Case Study Firm The carrier interviewed operates a large nationwide system with 1,800 terminals, 63,000 pickup and delivery trucks, 7,800 line-haul tractors, and 37,000 trailers. It picks up more than 11 million packages a day, of which 26,000 (0.2 percent) are declared by the shipper to contain hazardous materials required by regulations to be accompanied by special hazardous materials shipping papers. The average hazardous materials package size is 12 to 15 lb. The most common hazard classes are flammable liquids and corrosives, which together account for 70 percent of hazardous materials packages. The carrier refuses to carry certain hazard classes, including explosives A and B and radioactive II and III. The company moves its trailers by rail flatcar for most line-haul segments over 3 days' highway travel time. It also offers air package service, but this description deals only with ground operations. Most of its tractor-semitrailer mileage is with the twin-trailer configuration. The two twin trailers in a single combination together will carry about 2,500 packages, of which an average of perhaps half a dozen may contain hazardous materials. Thus it is possible that more than half of all the carrier's tractor-semitrailer combinations on the road contain at least some hazardous materials, although usually in very small quantities.
Shipper, Carrier, and Responder Information Capabilities 205 Steps in Handling Shipments At the time of pickup, the driver receives a pickup record listing all packages from the shipper and the shipper's name, package destina- tion, and weight. The pickup record stays at the originating terminal. This list and the address on the package are the only records of most shipments. There is no unique package identification number assigned by the carrier and no central file of packages. Transactions are virtually all on paper, and EDI is not used for transmission of any pickup or package records between shipper and carrier. Hazardous materials are the only shipments accompanied by a ship- ping document during transport. The shipper affixes a company form entitled "Hazardous Materials Shipping Paper" to each package for which a shipping paper is required by regulation and gives the pickup driver a certification form listing each hazardous package. At each stage in handling, a copy of the multipart shipping paper form is removed from the package and placed in an envelope in the truck cab, along with papers for all other hazardous materials packages in the vehicle. No manifest or comprehensive list of all hazardous materials packages on the vehicle is prepared. When a trailer is carried by rail, an envelope containing copies of the shipping paper for each hazardous materials package in the trailer is given to the railroad, and another envelope of papers is attached to the outside of the trailer door. The company is introducing an optional premium ground service in response to customer demands that they have the ability to track shipments in transit. A bar code containing a unique package identifier will be attached to the package at the first terminal to handle it. The bar code will be scanned three times: at the originating and destination terminal and at final delivery. A central computer file of premium service packages will be maintained, but this data base will not record every handling of the package or identify the vehicle carrying the package. This service uses an existing tracking system developed for the company's air package operation. Requirements for Participating in a Central Reporting System or Clearinghouse The company now has limited capacity to track shipments or to re- motely identify the contents of trailers. The premium service provides
206 HAZARDOUS MATERIALS SHIPMENT INFORMATION some tracking capability, but it would have to be substantially modified to record trailer identification numbers for each hazardous materials package. Package bar codes or bar codes added to the hazardous materials shipping documents would have to be scanned, and the identification number of the trailer being loaded would have to be entered, at each trailer loading bay. Any procedure would require manual entry of the hazardous materials shipping document into a data base. The company is working on further automation of package han- dling, which may reach the point of providing full package and vehicle tracking. A system is being tested in which an advanced bar code label is affixed to each package at its originating terminal. The labels are generated by voice entryâan employee reads the address on the pack- age aloud, and the label is printed. The initial application of the labels is to automate package sorting in terminalsâa robot would read the label and push the package onto the proper conveyer belt or bin. This application would not require that any central computer file of pack- ages or any record of the trailer carrying a package be maintained, but the system would make tracking much more feasible. The company also will soon have continuous in-vehicle communica- tion with its pickup and delivery trucks for automated electronic entry of pickup and delivery records. Drivers will have portable terminals for entry of the information on the pickup register, as well as more specific information about air freight and premium packages. One option the company could consider if it saw no practical way to comply with new hazardous materials tracking requirements would be to decline to handle such freight. Cost of Participating in a Central Reporting System or Clearinghouse To track hazardous materials packages by modifying the company's premium service would involve labor costs for special handling and substantial development and capital costs for modifying its current system and installing readers at new locations in its terminals. The company charges a fee of $6.00 per package for hazardous materials. It reports that the fee is based on an analysis of the costs attributable to the special handling and paperwork that these packages require. The fee for the premium service is only $0.75, reflecting the
Shipper, Carrier, and Responder Information Capabilities 207 economy of taking advantage of the preexisting capacity for air freight tracking to provide the ground service. RAI LROAD Company Description Compared with trucking, the rail industry is concentrated: seven sys- tems account for 80 percent of all freight revenues. The case study carrier interviewed is one of the major railroads. The railroads haul freight in cars they lease or own themselves, in cars owned or leased by shippers (e.g., the fleets of tank cars owned by chemical manufacturers), and (on flatcars) in highway trailers and intermodal containers owned by other transportation companies. Ex- cept for intermodal and some special services like carriage of new automobiles, nearly all rail freight is bulk, so most shipments are carload or multicarload quantities. Steps in Handling Shipments and Shipment Records Steps in the handling and record keeping for a typical shipment are as follows. The shipper submits a bill of lading to the company. Today about half are via EDI, and this fraction is expected to grow. Bills submitted by fax or paper are entered by clerks in the company's waybill data base. The record for each shipment includes the car number (there is an industry-standard car-numbering system) and hazardous materials information required by regulation. The shipper loads the car and notifies the railroad that it is ready to be picked up. A local train picks up the car, and the conductor on the train calls the central office to notify it that the pickup has been made. The shipper release of the loaded car as well as the time and location of all subsequent handlings of the car are recorded in the railroad's car location data base. Waybills are generated for empty cars in transit whose last load was a hazardous material, because they are considered to be carrying residue and therefore to require shipping papers according to the regulations. Car location records and waybills are maintained internally by the railroad and on the industrywide car-tracking system TRAIN II. This
208 - HAZARDOUS MATERIALS SHIPMENT INFORMATION system was developed for interrailroad communications: to manage equipment utilization, to handle interchange of cars between railroads, and for other purposes. The railroad makes hourly transmissions to TRAIN II of new waybills and updated car location records. In response to customer requests, the railroad can report the load status and location of any car and find the waybill for the car or find the car corresponding to any waybill, using information from TRAIN II for cars it has transferred to another railroad. Customers can query the car location data base directly to trace their shipments. Some shippers require regular periodic electronic reporting of shipment status. An industry-standard format for these communications exists, CLM (Car Location Message). Often this regular status reporting is handled by a third-party EDI network service. Car owners can query the railroad that picked up their car or subscribe to TRAIN II and search for their car there. Theoperators of TRAIN II are also developing services for other shippers. Requirements for Participating in a Central Reporting System or Clearinghouse With some exceptions the basic information and communication facili- ties that would be required to participate in a central reporting system or clearinghouse are in place. The company would incur development and operating costs for the new procedures required. Gaps in the system can arise in the case ' of some intermodal freight and because of time lags, sometimes of several hours, between car handlings and entry in the data base. The system contains waybills for containerized freight as it would for any bulk railcar shipment but does not contain waybills for small package trailers carried on flatcars. The waybill records in TRAIN II identify materials by an industry-standard code number (the Standard Transportation Commodity Code) only, not by shipping name. The railroad receives only paper records of hazardous materials contained in these trailers, which accompany the trailers on the train. The railroad's information system also contains records of the con- sist of each train, the list of all cars making up the train, which can be important for emergency responders. (TRAIN II does not provide access to consists.) The principal value for hazardous materials safety of the industrywide introduction of automatic electronic identification
Shipper, Carrier, and Responder Information Capabilities 209 transponders on all cars, which is now under way, may be that these devices will improve the accuracy of consists. LTL TRUCKING COMPANY Company Description The company is a major LTL carrier. In this segment of the trucking industry, companies specialize in carrying small shipments (averaging around 1,000 lb). Each firm operates a network of terminals. Shipments are picked up from shippers' facilities and hauled to a local terminal where they are reloaded onto trailers containing shipments headed in the same direction. A shipment may be unloaded and reloaded several times be- tween pickup and delivery. Networks have a hub-and-spoke pattern focusing on large regional consolidation terminals. The company operates about 600 terminals and 9,000 tractors and picks up 50,000 shipments each weekday. About 3 percent of ship- ments are hazardous materials, but because each tractor-semitrailer in intercity service is carrying 20 shipments on the average, about half of all tractor-semitrailers have at least one hazardous material on board. The mean size of a hazardous material shipment is 1,000 lb. By weight, flammable liquids are the most common hazard class (38 percent of tons); corrosives (26 percent) and flammable gases (13 percent) are the next most common. Shipment Handling and Record Keeping The company has an on-line shipment information system that allows identification of contents of trailers in intercity use. The shipper gives the driver of the pickup vehicle a paper bill of lading. At the local terminal, the bill of lading is manually entered in the shipment informa- tion data base and a bar coded waybill is generated. (In the iS percent of shipments for which shipping documents are received via EDI, manual entry is not necessary.) The waybill accompanies the shipment at all times. A loading dock worker scans the waybill bar code each time the shipment is placed on a dock or loaded on a trailer from the dock throughout the course of the shipment in the carrier's network. The carrier can search the shipment information data base by trailer number to obtain the bill of lading. A customer can search by waybill
210 HAZARDOUS MATERIALS SHIPMENT INFORMATION identification number to learn the location of a shipment, the next terminal it is scheduled to arrive at, the expected time of arrival, and the number of the trailer holding the shipment. Customers can receive information via ED! or telephone synthesized-voice response. The voice system handles nearly 300,000 inquiries per month. The principal gap in coverage of the shipment information system is that no record is made at the time of pickup or delivery. The bill of lading is not entered until the shipment arrives at the local terminal, and no centralized record is kept as the delivery vehicle unloads ship- ments at their destinations. Therefore, no central record exists of the contents of pickup and delivery vehicles. The pickup and delivery segments may account for 10 percent of the mileage a shipment travels between origin and destination. Other sources of incomplete coverage of hazardous materials arise from shipper failure to properly declare a hazardous material and from routine data entry errors or equipment failures. Requirements for Participating in a Central Reporting System or Clearinghouse Fr the intercity segment of a shipment's route, the company now has the computer data files that would be necessary for reporting haz- ardous contents of vehicles to a central reporting system or maintaining the company data base that would be required in a clearinghouse system. It would incur costs for development, communications, mainte- nance, and administration to participate in either arrangement. It would also incur costs for adopting some industrywide standard trailer numbering system, which probably would be necessary for the success of either arrangement. The most difficult and costly problem would be to create and operate a system to maintain central records of the contents of pickup and delivery vehicles. The lowest-cost method might be to equip pickup and delivery trucks with mobile communications devices and data entry terminals. Drivers would have to enter the critical bill of lading data at the point of pickup. This task could be simplified if shippers could be induced to transmit all hazardous materials bills of lading in advance via ED!. Trucking companies are beginning to experiment with systems having this capability, but the costs are still in the range of thousands of dollars per vehicle for hardware, in addition to communications and driver labor costs.
Shipper, Carrier, and Responder Information Capabilities 211 CHEMICAL MANUFACTURER Company Description The case study company is a major chemical manufacturer. It makes about 1.5 million hazardous materials shipments per year, 60 percent of all its product shipments. Products include every regulatory hazard class. Shipments include packaged materials as well as bulk carloads and truckloads. By volume, 90 percent of hazardous materials ship- ments are liquid. In terms of product volume, one-third of all shipments are by truck, and two-thirds are by rail and water. In terms of numbers, most shipments are by truck. One-third of all truck shipments are by the private fleet of a subsidiary; the remainder, including all LTL quan- tities, are by for-hire carrier. All rail shipments are in company-owned or leased cars. The company makes some international shipments in tank containers, and some of its domestic shipments handled by for- hire carriers are intermodal. Shipment Handling and Record Keeping The company's order entry system keeps track of all orders from receipt through shipping and billing. One function of the order entry system is to automatically generate a bill of lading for each shipment, which includes all required hazardous materials information. Shipments can- not be traced to vehicles in the system, because records of shipment and receipt are not entered at the time these events occur and no identifica- tion of the vehicle on which a shipment is made is entered. In day-to-day operations, there is no essential difference between the relationship of the company with its private carrier subsidiary and with the for-hire carriers it employs. The subsidiary does not have an auto- mated system in which the bill of lading could be identified on the basis of a trailer identification number. Deliveries are recorded when the driver calls the central office on a public telephone, but no computer entry of delivery is made. Most of the carrier's trailers are dedicated to carrying a single product. On this basis the carrier could determine what product would be in a specified trailer, but the carrier would have no certain way of determining whether a trailer that was in use was loaded or unloaded at a particular time.
212 HAZARDOUS MATERIALS SHIPMENT INFORMATION For rail shipments, all the company's cars are dedicated to particular products. To determine whether a car was loaded or unloaded, the company would have to refer to the railroad's car location information service (described in the railroad case study in this appendix). Requirements for Participating in a Central Reporting System or Clearinghouse It would be possible to develop procedures to record shipments as they were made so that a trailer or railcar identifying number could be attached to the bill of lading in the company's information system. The costs of such a system would probably include installation of new terminals and communication lines in shipping areas; new software systems; and labor costs for data entry, maintenance, and oversight and testing to maintain reliability. The company would have no way of keeping informed of deliveries to customers as they occur, except in cases where the carrier first devel- oped its own system to keep real-time records of all shipments it handled. The railroads already have this capability, and a few of the larger trucking companies are beginning to develop it. Therefore, the circumstances at this company, which are typical of many shippers, suggest that the most efficient arrangement for maintaining real-time records relating vehicle identification to contents would be to place record-keeping responsibility with the carriers rather than sharing it among carriers, shippers, and receivers. The shipper could facilitate the maintenance of the carrier information system by submitting orders for pickups via ED! so that the carrier would not have to manually enter waybill information. Electronic transmittal of shipping documents could be especially important if it reduced requirements for drivers to record and transmit shipment information. AIR FREIGHT CARRIER Company Description Air freight revenues of U.S. air carriers are about $7 billion annually; domestic and international traffic volume of U.S. carriers is 12 billion ton-mi. Freight is carried in the holds of passenger planes and in dedicated freighters. Air freight hazardous materials regulations spec-
Shipper, Carrier, and Responder Information Capabilities 213 ify limits on the quantities of materials that may be carried on one aircraft and forbid certain materials altogether. Restrictions are stron- ger on passenger aircraft than on all-cargo aircraft. The case study carrier is a passenger airline that carries freight in its passenger planes. Excluding mail, it handles 2.2 million shipments annually. Air cargo revenues are $500 million annually. Less than 0.5 percent of all shipments are declared hazardous by the shipper. The most common hazard class is flammable liquid. Shipment Handling and Record Keeping All shipments travel with an air waybill. In addition, shippers are required to provide a document declaring dangerous goods. Shipments are handled in the following steps: In response to a customer order, the shipment is picked up by a trucking company under contract to the air carrier and brought to the originating air freight terminal. At the terminal, papers, labels, and packaging of hazardous mate- rials shipments are inspected before acceptance. Hazardous materials shipments are charged an extra $7.50 per shipment for the special handling required. Waybill information is manually entered in the company's shipment-tracking system. The carrier does not make use of EDI for transmitting customer orders or documents. The tracking system is maintained as a customer service. The packaging and dangerous goods declaration identifying the material quantity are not entered as part of the shipment record. The pilot receives a manifest identifying haz- ardous materials on the flight. Real-time entries are made in the tracking system at each move- ment of the shipment in the terminal, when it is loaded on the plane, and when the flight departs. Similar entries are made at the destination. Requirements for Participating in a Central Reporting System or Clearinghouse The shipment-tracking system gives the company a current computer data base of the location of all shipments and the contents of each aircraft. The system does not cover ground pickup and delivery, which
214 HAZARDOUS MATERIALS SHIPMENT INFORMATION are performed by another company. To fulfill the intent of a central reporting system or clearinghouse system, the carrier would need to modify the data base to include the dangerous goods declaration. Some air carriers, mainly the smaller ones, do not have automated shipment-tracking systems, although such systems may be expected to become standard in the future. EMERGENCY RESPONDER CAPABILITIES Three elements of the emergency responder community are important components of any hazardous materials shipment information system: the first responders at the scene of an incident, hazardous materials emergency response teams, and public safety dispatch systems. This section describes the functions of these elements in response to a hazardous materials incident, the information systems they now use, and how they would interact with potential new hazardous materials information systems. First Responders Immediately after an incident, first responders must ascertain whether hazardous materials are involved (e.g., by looking for placards, reading papers, or speaking to the driver). If normal sources of information are inconclusive, first responders must take care not to expose themselves to injury while attempting to resolve the ambiguity. Once the likely presence of hazardous materials has been confirmed, most first re- sponders cordon off the area and seek expert assistance from special hazardous materials emergency response teams from the public or private sector. Currently, there is significant variability in the information systems that first responders possess. Nearly all emergency responders have mobile radios linking them to a dispatch center. A smaller number have cellular phones providing them with wider connectivity. A still smaller number are equipped with field facsimile capability. In some of the hazardous materials shipment information system concepts explored in this study, emergency responders would be re- quired to have more sophisticated communications facilities (e.g., cel- lular phone, portable facsimile with cellular phone, or lap-top personal
Shipper, Carrier, and Responder Information Capabilities 215 computers with modem and cellular phone). Given the current capa- bilities of first responders, some investment in telecommunications equipment will be necessary if any of these concepts is to be imple- mented widely. Hazardous Materials Emergency Response Teams Many municipalities and regional organizations have assembled teams with special equipment and training to cope with hazardous materials incidents. Generally, they are not the first emergency responders on the scene but respond if the presence of hazardous materials is suspected or confirmed. Such teams are called on to perform some or all of the following functions: determine whether hazardous materials are pres- ent at an incident; if hazardous materials are present, identify them and their physical and chemical properties (particularly under the prevailing conditions); estimate the risk to people, property, and the environment; formulate an emergency response action plan (to include recommendations for evacuations and decisions on the appropriate protective clothing); and implement actions to contain or ameliorate the effects of the incident. The following information system equipment is used by the Los Angeles hazardous materials team and is representative of the capa- bilities of hazardous materials teams in large municipalities. A team operator has access to a Macintosh SE computer, a cellular phone, and a fax machine. Two key applications are resident on the computer: CAMEO, an application that helps the emergency responder identify the chemicals involved and select the appropriate levels of protective gear, and ALOHA, which accepts meteorological input data (e.g., temperature and wind speed) to predict plume trajectory from a release. The hazardous materials trailer carries self-contained breathing ap- paratus and a range of protective clothing that the hazardous materials team can wear over normal fire gear. The protective clothing provides progressively enhanced protection to the emergency responder at the expense of reduced mobility, dexterity, and vision. The trailer also contains some limited equipment to support chemical analyses of unknown substances, including electronic gear to check for selected gases, indicators to run confirmatory tests for selected suspected agents, and a mini qualitative analysis laboratory. In addition, various
216 HAZARDOUS MATERIALS SHIPMENT INFORMATION reference sources are stored on the trailer, including chemistry texts, emergency response handbook, and ancillary information, such as data on sewer lines. Additional sources of information in the cab of the hazardous mate- rials emergency vehicle include mobile data terminals (MDTs), status boards (by which responders can tell the dispatcher whether they are en route or at the scene of an incident), and an automatic vehicle location display system (which uses dead reckoning to enable them to visualize their position on an electronic map). The hazardous materials emergency team can exchange information with a dispatcher in a public safety dispatch center via phone, facsimile, an MDT, or a status board. In addition, the team can telephone a transportation emergency center (e.g., CHEMTREC) to identify and contact a representative of the shipper and to secure appropriate Mate- rial Safety Data Sheets (which can be sent by facsimile). The sophistica- tion of the information system supporting the Los Angeles hazardous materials emergency response team is such that it would need little, if any, additional equipment to participate in proposed hazardous mate- rials shipment information system concepts. Public Safety Dispatch Centers Nearly all fire departments in the United States use public safety dispatch centers with basic communications capabilities. At a recent meeting of users and developers of computer-aided dispatch (CAD) systems, it was estimated that approximately 12 percent of public safety dispatch centers are substantially automated. The more sophisti- cated centers provide for greater incident capacity, more detailed sup- port information, and better control of resources. The centers generally support three broad functions: call taking, dispatching, and supervision. Call takers receive calls from the public, determine the nature of the event, enter the information into the computer system, and queue the incident to a dispatcher if a dispatch is necessary. Dispatchers assign units to incidents, follow activity for ongoing incidents, check that units are responding appropriately, and perform incident update actions.
Shipper, Carrier, and Responder Information Capabilities 217 Supervisors have the overall responsibility for the dispatch pro- cess. To that end, the supervisor monitors the use and availability of resources, executes move-ups to balance coverage, and dynamically adjusts parameters to respond to the dispatch load. There is wide variability in the computer and communications facilities supporting the functions of dispatch centers, as well as in availability of ancillary features such as geographic information systems, decision aids, and enhanced 911 with features such as automatic number identification (AN!) and automatic location iden- tification (AL!) displays. The capabilities being designed into the Los Angeles Fire Depart- ment (LAFD) CAD system, which is projected to be operational shortly, illustrate the equipment used by the most sophisticated CAD users. The objective is to field a paperless, highly redundant system that exploits commercial-off-the-shelf information systems. The system will support the operators in the following ways. The LAFD system contains decision aids to prompt the call taker in evaluating an incident and in assessing an appropriate response. In addition, LAFD has acquired and maintains a data base on the haz- ardous materials stored in fixed sites throughout the city. The system will alert the call taker to the most hazardous materials that are present in the vicinity of an incident. The call taker also receives AN!IALI on calls coming in on 911. The dispatcher has access to map and tabular data and can com- municate with field units via radio, MDTs, and automobile vehicle location systems. The system also contains algorithms that help the dispatcher make efficient and timely resource allocation decisions. The supervisor can access the displays and communications of any call taker or dispatcher. !nitially, public safety dispatch systems could be integrated into any potential hazardous materials shipment information system by using telephone or facsimile communications. Integrating CAD systems into any future hazardous materials shipment information system would be facilitated by development of system standards and protocols. With future developments in technology, systems could play an increasingly significant role in hazardous materials emergency re- sponse. For example, they could provide external assistance in mitigat-
218 HAZARDOUS MATERIALS SHIPMENT INFORMA11ON ing the effects of a hazardous materials incident by communicating with traffic control systems to reroute traffic, or they could actively track shipments of exceptionally hazardous material within their juris- diction to facilitate response in the event of an incident (e.g., via selective application of a much-simplified version of the Department of Energy's STARBASE system described in Appendix D).
Study Committee Biographical Information Barry M. Horowitz, Chairman, is President and Chief Executive Offi- cer of The MITRE Corporation. He has been with MITRE since 1969 and has been a member of the Board of Trustees since 1989. Before moving to MITRE he worked with the Bendix Corporation and Gen- eral Precision. He received his bachelor's degree from City College of New York and his master's and doctoral degrees from New York University. His professional activities and awards include the Air Force Exceptional Service Award; Gold Medal Award for Engineering pres- ented by NOVA Chapter, Armed Forces Communications and Elec- tronics Association; member of the board and former president of the Lexington-Concord (Massachusetts) Chapter of the Armed Forces Communications and Electronics Association; director of the Atlantic Council of the United States; consultant to the Defense Science Board, the Air Force Scientific Advisory Board, and the National Academy of Sciences; member of the Defense Communications Agency Scientific Advisory Group; Adjunct Research Fellow at the Center for Science and International Affairs, Harvard University; and Overseer of North- eastern University. Mark D. Abkowitz is Professor of Civil Engineering and Management of Technology at Vanderbilt University. Before coming to Vanderbilt, he was at Rensselaer Polytechnic Institute and worked at the Office of Technology Assessment to conduct an assessment of heavy truck safety. He received his doctoral degree from Massachusetts Institute of Technology. His research areas are hazardous materials transport, intelligent mapping, risk assessment, routing and logistics, emergency response and evacuation planning, and strategic planning in transpor- tation. He currently serves as Chairman of the TRB Committee on Transportation of Hazardous Materials. John C. Allen is Manager of the Transportation and Logistics Program Office at Battelle. He has worked at Battelle since 1983. He formerly worked with the Interstate Commerce Commission as a transportation economist and with the U.S. Department of Transportation, Re- search and Special Programs Administration as a policy analyst. He has managed and participated in numerous studies relating to haz- ardous materials transportation. He received a bachelor's degree from Western Maryland College and an M.B.A. from the University 219
220 HAZARDOUS MATERIALS SHIPMENT INFORMATION of Oregon. He is a member of TRB's Committee on Transportation of Hazardous Materials. Edward R. Chapman is Manager of Service Interruptions of Santa Fe Railway. From 1988 to 1992 he was Manager of Hazardous Materials. He has worked with the Santa Fe since 1977. He received his bachelor's degree from Texas A&M University. Randolph M. Doi is Vice President and Director for the IVHS Strategic Business Unit for Motorola. He has worked with Motorola since 1971 and has held his current position since 1988. He received an M.B.A. from Nova University in Florida and a B.S.E.E. from Purdue University. He was a member of TRB's Executive Committee from 1989 to 1992. Lee K. Foote was until 1992 Manager of Corporate Electronic Informa- tion Security at Du Pont. He joined Du Pont in 1965 and in 1989 was honored with Du Pont's Marketing Excellence Award for his group's Electronic Data Interchange (ED!) leadership. He received an M.B.A. from Columbia University and a bachelor's degree from Dartmouth College. He is a former chairman of the Chemical Industry ED! Commit- tee, which has the objective of implementing ED! within the chemical industry and between it and its customers and suppliers. He currently is President of the Southern Chester County Development Foundation. John R. Gunter is Vice President of Information Services and Market Plans for BellSouth Corporation in Atlanta. He has worked with Bell- South Corporation since 1964 and has held his current position since 1988. He received his bachelor's degree from the University of Missis- sippi. He is a member of the boards of directors of BellSouth Advanced Networks and BellSouth Cellular and serves as BellSouth's representa- tive for the Southern Technology Council of the Southern Growth Policies Board. Robert J. Hasken is Manager of Hazardous Materials Transportation for Corporate Environmental Programs with General Electric Company in Fairfield, Connecticut. He has been with General Electric since 1955. Dal L. Howard is the Assistant Fire Marshal for the Los Angeles City Fire Department and has held this position since 1991. In his 32-year career with the Department, he has served as the Communications and
Study Committe Biographical Information 221 Information Systems Division Commander; Supply and Maintenance Division Commander; Management Information Systems Section Commander; Planning Section Commander; and Battalion Com- mander, Bureau of Fire Suppression and Rescue. He is a graduate of Glendale College and Harvard University's John F. Kennedy School of Government Senior Executives Program. Gerald A. Long is President of Summit Information Systems, Inc., the information technology subsidiary of Roadway Express, Inc. He was formerly with Babcock and Wilcox, where his positions included Man- ager of Systems and Data Processing and manager of Materials Man- agement. He came to Roadway Express, Inc., in 1987 as director of Systems Development and started his current position in 1993. He received a bachelor's degree from St. John's University, New York. James M. Melius is Director of the Division of Occupational Health and Environmental Epidemiology for the State of New York Depart- ment of Health. He was formerly Associate Director for the Great Lakes Center for Occupational Safety and Health; and Chief of the Hazard Evaluations and Technical Assistance Branch as well as Direc- tor of the Division of Surveillance, Hazard Evaluations and Field Studies, for the National Institute for Occupational Safety and Health. He received his medical degree from the University of Illinois College of Medicine and his master's and bachelor's degrees from Brown Univer- sity. He is Chairman of the Medical Advisory Board for the Interna- tional Association of Fire Fighters. John B. Moran is Director of Occupational Safety and Health for the Laborers' Health and Safety Fund of North America. Previously, he has worked in safety and health related positions with the Environmental Protection Agency, the National Institute for Occupational Safety and Health, American Optical Corporation, GEOMET Technologies, and Laborers-AGC Education and Training Fund. He received his bache- lor's degree from the Illinois Institute of Technology. He has had numerous appointments by the Secretaries of Health and Human Services and Labor to national occupational safety and health advisory committees. Stephen B. Storment is Deputy Chief of Special Operations for the Phoenix Fire Department. He has been with the Phoenix Fire Depart-
222 HAZARDOUS MATERIALS SHIPMENT INFORMATION ment since 1969 and has held his current position since 1988. The Special Operations section is responsible for all hazardous materials and technical rescue training and emergency response. Dan L. Williams is Deputy Director of the Office of the State Fire Marshal, Illinois. He previously worked with the Federal Bureau of Investigation and the Illinois Emergency Services and Disaster Agency, and he was Executive Director of the Illinois Commission on Atomic Energy. He has worked for both the Legislative and Executive branches of state government since 1975, including the Illinois Department of Nuclear Safety and the Department of Transportation, Hazardous Materials Section. His current position includes responsibility for the divisions of Petroleum and Chemical Safety, Arson Investigation, and Boiler and Pressure Vessel Safety. He has also been the Emergency Response Liaison to the U.S. Department of Transportation. He re- ceived his bachelor's degree from Western Illinois University.
The Transportation Research Board is a unit of the National Research Council, which serves the National Academy of Sciences and the National Academy of Engineering. The Board's purpose is to stimulate research concerning the nature and performance of transportation systems, to disseminate the information pro- duced by the research, and to encourage the application of appropriate research findings. The Board's program is carried out by more than 300 committees, task forces, and panels composed of more than 3,700 administrators, engineers, social scientists, attorneys, educators, and others concerned with transportation; they serve without compensation. The program is supported by state transportation and highway departments, the modal administrations of the U.S. Department of Transportation, and other organizations and individuals interested in the develop- ment of transportation. The National Academy of Sciences is a private, nonprofit, self-perpetuating society of distinguished scholars engaged in scientific and engineering research, dedicated to the furtherance of science and technology and to their use for the general welfare. Upon the authority of the charter granted to it by the Congress in 1863, the Academy has a mandate that requires it to advise the federal government on scientific and technical matters. Dr. Frank Press is president of the National Academy of Sciences. The National Academy of Engineering was established in 1964, under the charter of the National Academy of Sciences, as a parallel organization of outstand- ing engineers. It is autonomous in its administration and in the selection of its members, sharing with the National Academy of Sciences the responsibility for advising the federal government. The National Academy of Engineering also sponsors engineering programs aimed at meeting national needs, encourages edu- cation and research, and recognizes the superior achievements of engineers. Dr. Robert M. White is president of the National Academy of Engineering. The Institute of Medicine was established in 1970 by the National Academy of Sciences to secure the services of eminent members of appropriate professions in the examination of policy matters pertaining to the health of the public. The Institute acts under the responsibility given to the National Academy of Sciences by its congressional charter to be an adviser to the federal government and, upon its own initiative, to identify issues of medical care, research, and education. Dr. Kenneth I. Shine is president of the Institute of Medicine. The National Research Council was organized by the National Academy of Sciences in 1916 to associate the broad community, of science and technology with the Academy's purpose of furthering knowledge and advising the federal govern- ment. Functioning in accordance with general policies determined by the Academy, the Council has become the principal operating agency of both the National Academy of Sciences and the National Academy of Engineering in providing services to the government, the public, and the scientific and engineering commu- nities. The Council is administered jointly by both the Academies and the Institute ofMedicine. Dr. Frank Press and Dr. Robert M. White are chairman and vice chairman, respectively, of the National Research Council.
ci) 2 Ui bi 0-1 TRANSPORTATION RESEARCH BOARD National Research Council 2101 Constitution Avenue, N.W. Washington, D.C. 20418 NON-PROHT ORG. U.S. POSTAGE PAID WASHINGTON, D.C. PERMIT NO. 8970 ADDRESS CORRECTION REQUESTED 100002105 1A 1A 9512 001 Robert M Snith Research & Asst Matis SuDvr Idaho Transportation Dept P 0 Box 7129 Boise ID 83707-1129 CID ci)
