A Concept for a National Freight Data Program: Special Report 276 (2003)

Demands on the U.S. transportation system continue to evolve in response to changing patterns of goods movement and passenger travel and heightened concerns about transportation security. In the case of freight, the growth of international trade, the shift from a manufacturing to a service economy, deregulation, and the advent of freight logistics have all resulted in changes in the nature and volumes of goods shipped and the origins and destinations of shipments. At the same time, growing congestion on the nation’s roads and at transportation hubs, such as ports and airports, not only inconveniences travelers but also threatens to undermine the reliable and timely movement of goods so critical to the national economy and quality of life.

The effectiveness and efficiency of the freight transportation system are heavily dependent on reliable data to inform a range of decisions at all levels of government and in the private sector about economic and infrastructure investments and policy issues. Data on goods movements are needed to identify and evaluate options for mitigating congestion, improve regional and global economic competitiveness, enable effective land use planning, inform investment and policy decisions about modal optimization, enhance transportation safety and security, identify transportation marketing opportunities, and reduce fuel consumption and improve air quality. While data alone cannot guarantee good decisions, informed choices are not possible without good data.

Data on goods movements are collected by federal agencies and other public- and private-sector entities that monitor or analyze transportation and trade activities on a regional, state, national, or international level.

Because these data collection efforts are not coordinated, the resulting data sets are of varying quality and reliability and provide an incomplete picture of the universe of freight movements. Furthermore, difficulties in combining data from the diverse sources limit the usefulness of current data sets for the purposes of freight transportation analyses. To remedy these deficiencies, a national freight data framework is needed.

A concept for a framework to guide the development of a national freight database and related data collection and synthesis activities is proposed in this report. This conceptual framework focuses on increasing the linkages between different sources of data and filling data gaps to develop a comprehensive source of timely and reliable data on freight flows. The national freight database aims to fulfill the major needs of a wide variety of users by capturing the important characteristics of freight movements— namely, shipment origin and destination; commodity characteristics, weight, and value; modes of shipment; routing and time of day; and vehicle or vessel type and configuration. The database also forms a foundation on which users can build their own more specialized data sets.

In its eight recommendations, the study committee offers the U.S. Department of Transportation and the Bureau of Transportation Statistics specific guidance on developing a multiyear program to implement the framework concept. In particular, the committee strongly recommends that a freight data advisory committee composed of stakeholders and experts drawn from both the public and the private sectors play a key role in guiding program development and implementation. The proposed initiative will require a sustained effort over many years and involve many technical and organizational challenges. The amount of data required is large, and some of the information needed by decision makers has not previously been collected in the United States. The committee highlights the development of innovative, low-cost methods for data collection and of procedures to protect the confidentiality of data providers as critical to a successful final outcome.

In 1998 the nation’s transportation system moved more than 15 billion tons of goods valued at more than $9 trillion. The Federal Highway Administration (FHWA) of the U.S. Department of Transportation (USDOT) estimates that by 2020, this volume will have grown by nearly 70 percent and the value of the goods moved will be almost $30 trillion (FHWA 2002). While these burgeoning commodity flows reflect growing economic activity, some of their impacts—particularly at the local level—are raising concerns about the challenges facing the nation’s transportation system. For example, congestion at ports, airports, and railheads and on streets, roadways, and highways slows not only the movement of freight, but also that of passenger traffic using the same transportation facilities.

Although concern about freight movements is not new, recent changes in the U.S. economy have highlighted a number of issues relating to freight transportation. Because of the growing importance of global markets and international trade, goods are being transported over longer distances than was the case 20 years ago. Furthermore, the patterns of goods movement have changed as a result of the growth of trade with Pacific Rim nations and among the North American Free Trade Agreement partners. Other changes in the business environment have also altered production, distribution, and logistics requirements. The shift from a manufacturing to a service economy, deregulation, and the advent of freight logistics have all resulted in changes in the nature and volumes of goods shipped and the origins and destinations of shipments (FHWA 2002). Much of the nation’s transportation infrastructure was built to accommodate patterns of

goods movement and passenger travel very different from those of today. As a result, the transportation system is increasingly challenged to provide the levels of efficiency and reliability in goods movement needed to ensure the international competitiveness of U.S. products and services and to sustain regional and local economies and quality of life.

The effectiveness and efficiency of the transportation system are heavily dependent on reliable information. In the case of freight movements, a range of decisions must be made at all levels of government and in the private sector about issues such as capacity utilization and potential shortfalls, congestion, safety, security, and environmental impacts. As noted in a report from the Bureau of Transportation Statistics (BTS), “good decisions require having the right information in the right form at the right time” (BTS 1998).

Recognizing the pressing need for reliable freight transportation data within the context of the changing U.S. economy and business environment, BTS was one of the sponsors of the conference “Data Needs in the Changing World of Logistics and Freight Transportation” held in Saratoga Springs, New York, in November 2001 (Meyburg and Mbwana 2002).¹ Conference participants representing a wide range of freight data interests concluded that freight transportation data are needed by a multitude of data users for varied reasons, ranging from broad policy issues to specific business logistics analyses. They also observed that existing freight data consist of a patchwork of data sets resulting from uncoordinated data collection efforts by diverse entities. Most of these efforts were not designed for the purposes of freight transportation analyses, and the resulting data are less than ideal for such applications. The most important action item identified by the Saratoga Springs conference sponsors and conferees was to develop a strategic freight data business plan to guide future data collection efforts. This plan “would identify all freight data users and their needs,” and on the basis of these needs “a national or international freight data architecture or framework would be developed” (Meyburg and Mbwana 2002, 23).

After the Saratoga Springs conference, BTS asked TRB to conduct a study to “recommend a framework for the development of national freight data.” The findings and recommendations of the committee formed to conduct this study are presented below. In accordance with its statement of task, the committee focused on developing a conceptual framework rather than a detailed data collection plan. The committee recognizes that the implementation of a national freight data framework will require sustained effort and funding over a number of years but believes that important benefits would accrue from streamlining data collection efforts, filling major data gaps, and harmonizing data sources for greater compatibility.

Transportation-related strategic planning in both the public and the private sectors relies on data on the movement of goods and people. Analysis of these data provides the information needed to inform policy and investment decisions. A 1992 report on data requirements for national transportation policy making noted that “without good data, decisions will be arbitrary, options overlooked, and solutions reactive” (TRB 1992, 5).

Policy issues affecting freight transportation include the identification of options for alleviating congestion in suburban and inner-city areas, the assessment of regulations governing the shipment of hazardous materials with a view to developing safer and more cost-effective approaches, and the identification of opportunities to enhance the security of goods movements without impeding the timely flow of commodities on which the nation’s economy depends. Decisions on all these issues require data on items such as the origin and destination of shipments, the commodities

moved, the modes of shipment, vehicle/vessel type, routing, and time of day. These data need to be provided in a timely fashion so that decisions can be taken on the basis of up-to-date information.

The private sector needs freight transportation data to identify underserved and emerging markets as well as potential efficiency improvements meriting investment. Data that help match loads to empty capacity are particularly important in allowing shippers to increase capacity utilization at very low marginal rates, thereby reducing shipping costs. Since much of the nation’s freight transportation infrastructure is privately owned and almost all freight is carried by private firms, private-sector investments have an important influence on the transportation system as a whole.

Many transportation analyses require freight data that are not available from any single source. Thus, it is frequently necessary to combine data from different sources. For example, carrier data from the Port Import Export Reporting Service database have been combined with shipper data from the U.S. Bureau of Customs and Border Protection to inform maritime infrastructure planning and analysis.

Even data sources covering all modes fall short in meeting many of the data needs of analysts and researchers. The Commodity Flow Survey (CFS), conducted every 5 years by BTS and the Census Bureau, aims to provide reasonably comprehensive information on the flow of goods by mode of transport within the United States. However, gaps in coverage limit the usefulness of the CFS data. For example, because the survey samples only domestic shipper establishments, it does not provide reliable data on the flow of goods into the country—data that are increasingly needed because of the growing importance of international trade to the U.S. economy.² Furthermore, a survey of shipper establishments alone cannot pro-

vide a complete picture of freight movements. For example, a shipper may specify a date and time by which certain goods need to arrive at their destination but may depend on the carrier to select an appropriate means of transportation. Consequently, the modal information from shippers captured by the CFS is incomplete in some cases. A further deficiency of the CFS is that the survey does not provide the level of geographic detail required by many state and metropolitan planners and engineers, who need to assign commodity and vehicle flows at least to corridors, if not to major highways and rail lines. The CFS also provides little information about unused, available existing capacity in the freight system. As a result of these limitations, the CFS data are often supplemented by data from other sources for the purposes of analysis and modeling.

The combination of data from different sources, often known as “data fusion,” is frequently problematic. Most existing sources of data on freight transportation were developed independently of each other to meet the needs of specific users. Thus, these sources vary considerably in terms of their modal coverage, data collection strategies, and data definitions. The possibility of using these disparate sources to populate a comprehensive national freight database raises concerns about the quality and comparability of the resulting combined data.

A further deficiency of existing sources of freight transportation data is that some of the information required by decision makers is simply not available. For example, informed efforts to alleviate highway congestion require data on routes traveled, time of day, and the types of trucks and commodities caught in congestion—data that are rarely collected, at least in the United States.

Both the committee’s discussions with users and the personal experience of individual members revealed a sense of frustration with existing freight data. The disjointed array of data sources is cumbersome and difficult to use, lacking in geographic detail, and notably deficient in covering increasingly important motor carrier flows. Several users also expressed concern about the unnecessary burden on data providers, who may be asked to provide similar data to different organizations—sometimes in different formats. This heavy respondent burden is likely to hinder efforts to gather quality data.

The committee concluded that the present patchwork of uncoordinated and incomplete freight data sources needs to be reengineered in

the context of a national freight data framework that provides for a more integrated approach to freight data collection and synthesis. In the committee’s view, it would be impractical—and prohibitively expensive—to subsume all existing data sources into a new monolithic freight data framework. Furthermore, existing data requirements and the need for continuity of data to inform trend analyses limit the possibilities for “wiping the slate clean” and initiating an entirely new set of data collection programs. Therefore, the committee envisages a framework with a flexible, modular structure that leverages existing data sources as far as possible and avoids costly new data collection initiatives unless they are needed to fill data gaps.

A national freight data framework—such as that shown schematically in Figure F&R-1—that facilitates data fusion and fills data gaps would aid in developing a comprehensive picture of freight flows. Such a picture offers opportunities to enhance the security of goods movement by identifying vulnerabilities. For example, data on routing and time of day for bulk shipments of hazardous materials could be used in identifying high-risk scenarios and targeting appropriate security measures. In the longer term, an improved understanding of normal freight flows would also provide a baseline against which to identify anomalies linked to possible acts of terrorism. A recent report on countering terrorism identified an understanding of normal patterns of transportation activity and behavior as a key research need for transportation security (NRC 2002).

The improved freight data resulting from the framework approach could also increase the international competitiveness of U.S. products and services through more effective use of and improvements in the nation’s transportation system. In addition, these data could help sustain the current strengths and stimulate the development of regional and local economies through informed decisions that take appropriate account of freight transportation needs and opportunities.

Figure F&R-1 Proposed framework of a national freight data program. [BTS = Bureau of Transportation Statistics; O-D = origin–destination; DHS = Department of Homeland Security; MPO = metropolitan planning organization; state DOT = state department of transportation. Source: Adapted from a paper prepared for the committee by R. Donnelly (Appendix A).]

Although many of the issues in freight transportation are well known, comprehensive high-quality data may be useful in pinpointing underlying causes and prioritizing policy and investment decisions. Such data may also enable research aimed at solving freight transportation problems— research that cannot currently be undertaken because the necessary data are not available.

FHWA recently completed a 3-year project to develop the Freight Analysis Framework (FAF)—a national-level policy analysis tool (FHWA 2002). The FAF, which is based on a composite of many databases, has played, and continues to play, an important role in raising awareness of freight issues. It is also a “point of departure for further examination of policies, programs, and initiatives that might be undertaken by decision

makers at all levels of government” (FHWA 2002, 2). In many instances, such examination will require the kinds of data that a national freight data framework would provide.

• Origin and destination;

• Commodity characteristics, weight, and value;

• Modes of shipment;

• Routing and time of day; and

• Vehicle/vessel type and configuration.

The committee’s discussions with data users indicated that providing all the data needed for all applications would be beyond the scope of any national initiative. In the committee’s judgment, a national freight data framework program needs to focus on populating a national freight database that provides robust data to meet basic user requirements. Providing options to enhance the data set for more specialized purposes would be extremely beneficial in meeting a wider range of user needs. On the basis of its discussions with users, review of the literature,³ and the experience of individual members, the committee identified the data items listed above as particularly important for a wide variety of applications. Not all the items would be used for every analysis; users would “pick and choose” the data they require.

No source, public or private, currently provides reliable data on all the items listed in Finding 4. Only the CFS and the related Transearch database from Reebie Associates even recognize the need to provide a complete description of freight flows in the United States, and both have important deficiencies. As discussed under Finding 2, the CFS has inherent limitations because it samples only domestic shipper establish-

ments. The survey also suffers from gaps in industry coverage because it excludes transportation and service establishments and most retail establishments, as well as agricultural shipments from the farm to the first point of assembly. While the Transearch database attempts to fill some of the gaps in the CFS, its coverage is also incomplete because of limitations in the data sources available for input. In addition, the Transearch database is generated through proprietary methods, and information about data reliability is not reported.

Many of the data items listed above are already available from different sources, but, as noted earlier, combining data from different sources is challenging. Furthermore, there are gaps in existing data that are extremely difficult to fill other than by data synthesis. In terms of modal coverage, the most significant gap is in motor carrier flows—an important growth area that is not well covered by any existing data collection efforts. Of the data items listed in Finding 4, routing and time of day are sparsely covered by current data sets, even though they are important for assessing congestion mitigation strategies, evaluating system capacity, and ensuring the security of shipments.

No single organization by itself has the resources and expertise necessary to develop and implement a national freight data framework. The interest and cooperation of a range of public- and private-sector organizations will be essential to the overall success of the proposed framework initiative. The participation of industry will be particularly important given that almost all freight in the United States is carried by private firms.

In addition to a team effort involving public- and private-sector participants, the framework initiative will require strong leadership to coordinate the data collection activities of diverse entities within the context of an overall strategy. Given the national significance of the proposed framework, the diffuse nature of some of its “public good” benefits, and the need for a systemwide approach involving all levels of government

and the private sector, the committee concluded that an organization within the federal government would be best positioned to assume the leadership role.

The following recommendations address the initial programmatic and technical steps required to move forward with the implementation of a national freight data program.

In the committee’s view, USDOT has the knowledge and expertise needed to lead the development and implementation of a national freight data framework. Furthermore, the ultimate goal of the framework— namely, providing a comprehensive picture of freight flows in the United States—is consistent with the department’s mission of ensuring a transportation system that meets vital national interests and enhances the quality of life of the American people.⁴

The modal administrations within USDOT are already involved in mode-specific data collection programs for operating and administrative purposes. Their expertise is complemented by that of BTS in multimodal freight surveys (the CFS) and in statistics and survey methodology in general. The experience of FHWA in developing the FAF could also be helpful in implementing the proposed national freight data framework initiative.

Although USDOT has considerable knowledge and expertise in the area of freight data, the national framework initiative is too broad in scope to succeed without the support and involvement of the wider freight data community. Therefore, the committee envisages that the development and implementation of the framework will be guided by a freight data advisory committee of stakeholders and experts. The membership of this advisory committee should reflect the broad spectrum of freight data users and providers and should include representatives of federal, state, and local jurisdictions as well as a range of private-sector stakeholders. The latter group should include consulting companies, representatives of different modes of transportation (air, maritime, pipeline, railroad, trucking), shippers and receivers, third-party logistics companies, and academic researchers. Since national defense activities, such as those of the Army’s Military Traffic Management Command, could benefit from improved freight flow data, the committee suggests that the advisory committee membership include an expert in defense logistics.

The legislation authorizing the establishment of BTS requires the agency’s statistics to support transportation decision making by all levels of government, transportation-related associations, private businesses, and consumers [49 U.S.C. 111(c)(7)]. Furthermore, as a federal statistical agency, BTS is charged with the continual development of more useful data “to provide information that is accurate, timely, and relevant for changing public policy needs” (NRC 2001). In light of these requirements and the agency’s experience and expertise in survey methodology and statistics, the committee considers it appropriate for BTS to play a major role in the freight data framework initiative.

Details of the BTS role will require further investigation as plans are developed for implementing the framework. For example, it could be argued that BTS should assume full responsibility for developing and maintaining a national freight database because of the need for transparency

(no proprietary data or methods)⁵ and the likely need to work with confidential microdata. On the other hand, extensive expertise in data fusion and manipulation exists in the private sector. For this reason, some would argue that BTS should focus on data collection and leave the development of a national freight database to others. More detailed definition of the tasks involved in database development—and other aspects of framework implementation—will be needed before appropriate roles for various public- and private-sector groups can be assigned on an informed basis.

• Examination of the relative merits of different kinds of surveys (carrier, distributor, shipper, receiver)—and combinations thereof—for gathering the types of data users require;

• Identification and preliminary assessment of opportunities to use data sources other than surveys—for example, electronic data interchange (EDI) and intelligent transportation system (ITS) data— to collect data more cost-effectively, fill data gaps, and reduce respondent burden;

• Identification of opportunities for facilitating the integration of data from different sources; and

• Investigation of techniques to assist a range of users in combining data from supplemental sources (urban truck surveys, Customs data, security data, etc.) with national freight data to meet their individual requirements.

The committee envisages that implementation of the framework will involve the development of a freight data road map that builds on the con-

cepts presented in this report. Such a road map will need to assess the state of various data collection, synthesis, and fusion techniques; identify opportunities for coordinating data collection efforts and filling data gaps; and identify promising areas for research and innovation to support implementation of the framework.

Assessment of the strengths and weaknesses of current data sources and of different types of surveys will be a prerequisite to setting priorities for further investigation and development. The committee does not envisage that all possible approaches will be pursued concurrently or that equal effort will be devoted to each. The freight data advisory committee will need to weigh the costs of obtaining data against the potential benefits of making the most effective use of limited resources.

Because of the large amounts of data required and the relatively high costs of surveys, implementation of the framework will need to take advantage of nonsurvey data streams. Opportunities to use low-cost passive data collection appear particularly promising. For example, ITS roadway surveillance data, which are generated continuously and at a very detailed level, could be used for congestion monitoring and intermodal facilities planning, over and above their use in real-time control strategies. Preliminary requirements have already been identified for an ITS Archived Data User Service to support a wide range of stakeholder activities (Margiotta 1998).

Modifications of current data collection efforts that facilitate the integration of data from more than one survey or of survey data with ITS or EDI data could be a particularly cost-effective way of developing more useful freight data. As noted in a recent National Research Council report, “when separate datasets are collected and analyzed in such a manner that they may be used together, the value of the resulting information and the efficiency of obtaining it can be greatly enhanced” (NRC 2001, 7). The committee anticipates that a major strength of the framework will be the provision of opportunities for states, metropolitan planning organizations (MPOs), railroads, transportation-related associations, and the like to build on a national freight database in developing their own data sets. To facilitate the combination of specialized and national data, the framework will need to provide standard survey methodologies and examples of their use, together with recommended best practices for activities such as data collection and processing. The committee envisages a leadership

role for BTS, as a statistical agency within USDOT, in developing standard methodologies and encouraging the use of best practices in freight transportation surveys.

• Focus on providing real data—as opposed to imputed or synthesized data—when possible.

• Ensure that data are sufficiently timely to meet the needs of users.

• Ensure that data collection and synthesis methods are appropriately documented so that data users can assess the resulting data quality and reliability.

• Encourage the use of compatible data elements to facilitate the combination of data from different sources.

• Incorporate mechanisms that encourage continuous feedback from users and subsequent refinement of the framework.

On the basis of its discussions with users, review of the outcomes of the Saratoga Springs meeting (Meyburg and Mbwana 2002), and the experience of individual members, the committee understands that the provision of real (as opposed to synthesized) data in a timely fashion, together with information on data quality and reliability, is important to users. These user expectations will need to be taken into account in developing and implementing the framework and in populating the national freight database. In particular, a decision by the freight data advisory committee concerning the level of geographic detail to be provided by the national freight database will involve balancing the needs of different users (federal government agencies, state departments of transportation, MPOs, consulting companies, academic researchers, etc.) for data on international, national, state-to-state, and intrastate goods movements and the costs of collecting, processing, and disseminating such data in accordance with requirements for timeliness, reliability, and the like.

While a national freight database cannot provide all data for all users, it can form a basis on which to build by facilitating linkages to other data sets. One of the major deficiencies of current freight data is the lack of harmonization among different databases. In the committee’s view, it is

imperative to move as rapidly as possible toward a more integrated approach that eliminates unlinked data “silos.” The use of compatible data elements, standard survey methodologies, and other techniques for facilitating data fusion will be essential to the successful implementation of the national freight data framework.

The committee anticipates that the content and detailed structure of the framework will evolve over time to reflect research findings, new opportunities for data collection, and practical experience with different data sources. Feedback from users will be essential in guiding this evolutionary process, and the freight data advisory committee will need to establish mechanisms for encouraging interactive development and implementation of the framework.

• Explaining clearly why data are being collected and for what purposes they will be used,

• Avoiding overly burdensome reporting requirements, and

• Respecting the imperative to maintain data confidentiality.

New data providers⁶ will need to see some payoff for supplying data, while current providers will need encouragement to expand the scope of data they supply or adapt to new data collection methods. In many instances, diffuse “public good” benefits are unlikely to constitute a particularly attractive incentive for participation in data collection programs. The promise of specific benefits, such as the availability of data on empty movements to help carriers increase load factors, are likely to be more persuasive.

The need to maintain the confidentiality of individual firms is a potentially fatal flaw that must be addressed early in the framework development

process. Data providers will not participate in framework activities if they see any risk of their competitors gaining access to commercially confidential information. Thus, in seeking to encourage private-sector participation, USDOT/BTS will need to recognize that much of the proprietary data collected for legitimate business planning and investment purposes cannot readily be converted into public use data.

A recent National Research Council report notes that federal statistical agencies need to treat data providers fairly. In addition to adopting policies and procedures for maintaining data confidentiality, agencies need to seek the advice of respondents in designing data collection procedures and determining data products (NRC 2001, 10). Such practices appear particularly pertinent in the context of efforts to develop and implement a national freight data framework.

The International Trade Data System (ITDS) project, which seeks to streamline the collection and dissemination of international trade data, may provide helpful insights for implementing the national freight data framework. In contrast to current practice, which often involves multiple reporting requirements, traders will submit standard electronic data only once. ITDS will then distribute these data to federal agencies on a “need-to-know” basis; each agency will receive only information relevant to its mission.⁷ Such standardized data collection could help reduce the burden on providers of freight data, and a system of strict controls on data access could allay the concerns of these providers about inadvertent release of commercially sensitive information.

The willingness of various groups in both the public and the private sectors to participate in framework implementation will largely depend on their perceptions of how useful the resulting data are likely to be. To encourage broad participation, USDOT/BTS will need to understand the

perspectives of different stakeholders and present the anticipated benefits of the framework accordingly. FHWA’s FAF program involved extensive outreach to freight stakeholders for the purposes of improving understanding of the nature of freight movements, identifying challenges to improving freight productivity and security, and developing strategies to increase freight productivity (FHWA 2002). Implementation of the national freight data framework is likely to involve comparable outreach activities to explain the value of better data in addressing various freight issues, solicit suggestions for framework development and implementation, and encourage participation in the framework initiative.

Some form of institutional structure, such as a program office, will be needed to coordinate activities within a national freight data framework program, support the freight data advisory committee, and provide a focal point for the framework initiative. A framework program office would also be responsible for facilitating continuous feedback and refinement of the framework, identifying emerging data collection opportunities and encouraging related research investigations, and ensuring the sustainability of data collection activities so that the national freight database can be expanded and updated. Over time, the focus of program activities is likely to shift from feasibility studies and concept development through implementation to updating and maintenance to ensure long-term viability. The institutional program structure will require sufficient flexibility to accommodate this evolutionary process.

BTS Bureau of Transportation Statistics

FHWA Federal Highway Administration

NRC National Research Council

TRB Transportation Research Board

BTS. 1998. Transportation Statistics Beyond ISTEA: Critical Gaps and Strategic Responses. BTS98-A-01. U.S. Department of Transportation, Washington, D.C.

FHWA. 2002. The Freight Story: A National Perspective on Enhancing Freight Transportation. FHWA-OP-03-004. U.S. Department of Transportation, Washington, D.C. www.ops.fhwa.dot.gov/freight/publications/freight%20story/freight.pdf.

Margiotta, R. 1998. ITS as a Data Resource: Preliminary Requirements for a User Service. Office of Highway Policy Information, Federal Highway Administration. www.fhwa.dot.gov/ohim/its/itspage.htm.

Meyburg, A. H., and J. R. Mbwana (eds.). 2002. Conference Synthesis: Data Needs in the Changing World of Logistics and Freight Transportation. New York State Department of Transportation, Albany. www.dot.state.ny.us/ttss/conference/synthesis.pdf.

NRC. 2001. Principles and Practices for a Federal Statistical Agency, 2nd ed. National Academy Press, Washington, D.C.

NRC. 2002. Making the Nation Safer: The Role of Science and Technology in Countering Terrorism. National Academy Press, Washington, D.C.

TRB. 1992. Special Report 234: Data for Decisions: Requirements for National Transportation Policy Making. National Research Council, Washington, D.C.