Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 2


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
1 SUMMARY Improving Pedestrian and Motorist Safety Along Light Rail Alignments Introduction TCRP Project A-30's goal was to provide guidelines for improving pedestrian and motorist safety along light rail transit (LRT) alignments. The project had three central objectives: To develop a framework or template for collecting data for assessing pedestrian and motorist safety along LRT alignments, To identify and summarize pedestrian and motorist behavior, and To document best practices for improving pedestrian and motorist safety along LRT alignments. To accomplish these objectives, a wide range of tasks was undertaken, and the project was divided into two phases. In Phase I, the project team undertook the following tasks: Literature review of LRT safety issues, safety measures, devices, practices, and new tech- nologies relevant to LRT alignments. Survey of North American LRT agencies. Preliminary round of assembly and analysis of basic LRT crash data obtained from fed- eral, state, and local transit agencies. Suitable data were difficult to obtain. Summary of all information gathered, and preparation of Phase 2 work plan. In Phase II, the team undertook the following tasks: Consultation with representatives of the Federal Transit Administration (FTA), State Safety Oversight (SSO) agencies, and local LRT operating agencies. Follow-up requests for data at the agency, state, and national levels. Site visits to selected LRT agencies. Five agencies were visited: Utah Transit Authority, Salt Lake City, Utah; Metro Transit, Minneapolis, Minnesota; Hudson-Bergen Light Rail Line, Hudson County, New Jersey; San Francisco Municipal Railroad, San Francisco, California; and Santa Clara Valley Transportation Authority, Santa Clara County, California. The five site visits included system observation, safety workshops, and stakeholder consultation. The visits allowed for the collection of further information about data collection, collision records, the use of the Manual on Uniform Traffic Control Devices (MUTCD), and LRT safety issues. The visits provided valuable opportunities for detailed discussion of the effec- tiveness of treatments, risk assessment, data gathering, and data processing. Further review of LRT safety data, including: compilation and analysis of newly available data, review of data collection and storage procedures, and development of recommen- dations concerning data collection and processing.

OCR for page 1
2 Creation of detailed catalog of LRT-related safety treatments (characteristics, examples of installation, safety effectiveness) Development of a risk assessment methodology. The second phase focused on successfully meeting project objective requirements that remained after Phase I and producing this final report, which documents all the project's activities, findings, and recommendations. The first objective listed above (developing a framework for collecting LRT safety data) was the most straightforward. The second objective (pedestrian and motorist behavior) and the third objective (best practices for improving pedestrian and motorist safety along LRT alignments) presented major challenges. In particular, the detailed research statement called for a review of the effectiveness of treatments based on experience to date. The project team examined the collision data gathered as part of the project and conducted an extensive litera- ture review, but found very little statistically significant information. The problem was ampli- fied by the quantity and quality problems identified in the National Transit Database (NTD) and by the paucity of data available directly from local transit agencies. In response to these limitations, the project team took a dual approach: assessing treatments quantitatively where quantitative data were available, and collecting and summarizing qualitative and anecdotal information where quantitative data were not available. This summary is divided into the following sections: State of the practice methodology and summary, LRT safety data available from local transit agencies, SSOs, and the NTD, Safety issues and their treatment, Catalog of LRT safety treatments, Risk analysis methodology for LRT, and Review of the accident data collection process. State of the Practice Methodology and Summary The number of LRT systems in North America is growing, and existing systems are expanding. There is great interest in further guidance on addressing safety issues. Agencies have developed their own strategies for improving safety on the light rail alignment, but there is no clearinghouse to share this information. The literature review, survey of agencies, telephone consultation with agencies and SSOs, and observations and additional informa- tion obtained during the site visits enabled the project team to identify and summarize the most important safety issues cited by agencies. A long list of important safety-related problems emerged: pedestrian jaywalking and trespassing, safe station access, illegal turns, traffic control violations, gate violations, conflict- ing signals and signs, poor signal clearance, best ways to provide traffic signal pre-emption, pinch points on platforms, sideswipes, lack of information for the light rail vehicle (LRV) operator, and the public's lack of respect for LRVs. The top concerns share common causes. Collisions occur when people cross into the alignment because of inattention, confusion, or purposeful noncompliance, or when operators fail to follow procedure and behave in unex- pected ways. These points may seem obvious, but agencies must make a distinction between preventing risky behavior due to inattention and confusion, and preventing risky behavior due to purposeful noncompliance. Understanding the underlying reasons for collisions on LRT alignments is essential to designing appropriate treatments. Agencies address safety concerns in two ways: physical measures, and education and enforce- ment programs. Physical improvements are designed to keep motorists, pedestrians, and

OCR for page 1
3 cyclists out of the alignment, or to give them information about the LRT alignment. Physical measures may be active or passive. Active measures respond to an approaching LRV. Passive measures are fixed. Education and enforcement programs may focus on operators or the pub- lic. Public programs are typically used to increase awareness about safety near LRT alignments. LRT Safety Data Available from Local Transit Agencies, SSOs, and the NTD In the United States, LRT collision data are collected and stored at three levels of transit administration. As part of the data collection portion of the project, the project team identi- fied the purpose of storing collision data at each level, identified the data collected, assessed the quality of the data collected, identified any potential issues or deficiencies, and, where pos- sible, analyzed the data to determine root causes and contributing factors for LRT collisions. The three levels of transit administration that collect and store incident reports are the FTA (for the NTD), the SSO agencies, and local transit agencies. The primary source of all data is onsite investigations using some type of collision report form. Normally, the local agency conducts the incident investigation, but the SSO is mandated either to conduct an investigation or to ensure that the local agency is doing so. The SSO reviews the findings of the local agency's incident report, and accepts or dissents. If the SSO dissents, the SSO may conduct its own review, or may ask the agency to revise the agency's review. The local tran- sit agency must also submit monthly reports on incidents meeting set criteria to the NTD. SSOs make an annual submission to the NTD. Figure 1 shows the transfer of incident data from the scene of the incident across the three levels of transit administration. The project team received and analyzed data from a number of sources: The National Transit Databases from the FTA for each year from 2002 to 2007, including all reported safety and security incidents, both "reportable" (formerly "major") incidents, and "non-reportable" (formerly "non-major") incidents; One SSO database, provided by the California Public Utilities Commission (CPUC); and Extractions of varying levels of detail from the databases of eight local LRT agencies. Safety Issues and Their Treatment Safety analysts use tools such as before-and-after studies to determine the most appropri- ate treatments to address safety issues. One method of addressing safety concerns is root cause analysis (RCA). RCA is a formal process that distinguishes contributing factors from root S&S-40 Reportable Incident Form within 30 days of incident Annual Submission S&S-50 Safety and Security Summary Form monthly NTD SSO Transfer of Final Investigation Report Local Incident Investigation Incident Investigation Incident Figure 1. Transit incident data transfer across levels of transit administration.

OCR for page 1
4 causes. Contributing factors influence the occurrence or severity of the collision, but their elimination would not have prevented a collision from occurring. RCA goes beyond con- tributing factors to stress the importance of identifying root causes: the underlying flaws or problems that lead to an undesirable outcome. Safety treatments should attempt to identify and mitigate the fundamental weaknesses of the system to prevent incidents from recurring. It is difficult to determine the root causes of LRT collisions statistically. It is also difficult to determine the effectiveness of LRT safety treatments statistically. Statistical problems arise because collisions (and especially specific types of collisions) are both rare and random events. Practical problems arise because of the inadequacies of much of the LRT data avail- able. Data collection practices need to be expanded and standardized to create a central, national database that will be useful for research studies. Because LRT collisions are rare and random, some types of statistical studies may never be meaningful, even with improved data collection. Proxy measures can, however, provide a useful substitute for collisions and can be used to make a quantitative assessment of the effectiveness of safety measures. Proxy measures could also be used in RCA. Proxy measures can have the advantage that they allow agencies to identify and treat safety hazards before there is an incident. Suggested proxy measures include: Operator reports, Risky behavior, Emergency braking records, Insurance and non-recoverable cost records, and Customer complaints. The project team analyzed the data available and considered the information obtained from agencies during the site visits. Four strategies are recommended to address safety issues: Give responsibility to the operators; Increase motorist, pedestrian, and cyclist awareness by providing active, appropriate information for them to act upon at each LRT location; Educate the public and increase awareness of risks in the LRT alignment; and Separate LRT space from the space occupied by other modes, using physical barriers, envi- ronmental cues, or traffic control. Catalog of LRT Safety Treatments A wide range of specific safety treatments is available and used along LRT alignments. Although little quantitative information is available, agencies have accumulated significant anecdotal information about the implementation and effectiveness of many treatments. The project team assembled this information, and any available quantitative information, into an easy-to-use, highly accessible catalog of LRT safety treatments. The catalog includes details of 31 LRT treatments. Where possible, photos of actual examples of the treatments are included. The catalog is designed to be used as a stand-alone resource. It is presented as Appendix A of this report. The 31 treatments are divided into 7 categories. The treatments and categories are: 1. Signals and active warnings a) Signal priority b) Transit signal pre-emption

OCR for page 1
5 c) Audible crossing warning devices d) Constant warning time systems e) Pre-signals f) Flashing light signals g) Limits on downtime of gates h) On-vehicle audible warning devices--automatic and LRVoperator-activated i) Illuminated, active, in-pavement marking systems j) Blank out signs k) Pedestrian signals 2. Signs a) Stop and yield signs b) Retroreflective advance warning signs c) Flashing train-approaching warning signs d) Gate crossing status indication signals 3. Second train approaching treatments a) Second train signals and active signs b) Second train warning signs 4. Gates a) Pedestrian automatic gates b) Four-quadrant gates 5. Pedestrians a) Pedestrian fencing/landscaping b) Offset (or Z) pedestrian crossings c) Pedestrian swing gates 6. Channelization/markings a) Pavement marking, texturing, and striping b) Quick curbs c) Rumble strips d) Channelizations e) Illumination of crossings 7. Education and enforcement a) Photo enforcement b) Enforcement c) Education outreach programs d) CCTV/video recording Risk Analysis Methodology for LRT The project team approached the task of developing a risk analysis methodology for LRT alignments by: Reviewing the established concept and principles of road safety audits, Consulting the FTA's "Hazard Analysis Guidelines for Transit Projects" standard, and Examining the process of safety issue identification in a number of recent sources from North America and the United Kingdom. A safety issue checklist for LRT agencies was then developed. The FHWA Road Safety Audit Guidelines document explains safety audit principles and shows how road safety audits are adapted to take into account the stage in the life-cycle of the road (from preliminary design to detail design and in-use stages). The document is a

OCR for page 1
6 good example of a road safety audit publication and is widely used in the road design and management field. The basic principles of road safety audits emphasize the importance of considering safety from the perspective of all possible users of the roadway. Users include pedestrians, cyclists, private motorists, and commercial operators. Their trip purposes vary (commuting, tourism, commercial vehicle operation, etc.) and their level of safety awareness and knowl- edge varies. During the audit, an independent multi-disciplinary team also examines all aspects of the road design, identifies potential risks, and suggests mitigation measures. The audit team's findings are then discussed with the road design team, and the most appropri- ate course of action is developed. The mitigation options available will vary greatly depend- ing on the problems and circumstances identified, the cost, and the stage of design and/or operation of the facility. The FHWA guidelines and two other widely used North American safety audit guideline documents were used to summarize the steps involved in conducting a safety audit: Select the safety audit team, Provide background information to the safety audit team, Conduct a pre-audit meeting to review project information, Assess/analyze background information, Perform site inspections under various conditions, Prepare and submit safety audit report, Conduct safety audit completion meeting, Prepare formal response (completed by project owner/design team), and Incorporate safety audit findings into project (where appropriate). The FTA's Hazard Analysis Guidelines for Transit Projects recommends conducting haz- ard identification, assessment, and resolution using Military Standard 882D (MIL-STD- 882D), the U.S. Department of Defense (DoD) Standard Practice for System Safety. This standard describes the process of examining and resolving safety issues and describes the requirements for developing and implementing a system safety program. During the site vis- its, a number of local LRT staff mentioned that the FTA Guidelines are a commonly used reference for safety management and a commonly used basis for safety problem resolution. The principles and approach adopted in road safety audits and by the FTA and others were used to develop a draft LRT Alignment Risk Assessment Checklist. The emphasis was on usability rather than including every detail of every possible issue because it was important to keep the checklist to a reasonable length. The draft checklist was presented and discussed at the safety workshops conducted during the site visits. It was then revised, and the final version is presented in this report. Review of the Accident Data Collection Process A key task was the development of recommendations for improving the collision data col- lection process at the local and national levels. LRT agencies, SSOs, and the FTA and its NTD are all involved in data collection, dissemination, and analysis. Improving the quantity and quality of the data available is an important part of facilitating the statistical analysis of LRT safety data at the national level. Collision data collection forms and methods were requested from LRT agencies. It proved difficult to obtain material from many agencies, but 11 agencies provided material. The proj- ect team conducted a detailed review of the forms provided. The review considered data quality, data consistency, gaps and redundancies in the data, and problems that arose when

OCR for page 1
7 incidents were selected and transferred from one agency to another. The analysis provided the basis for a set of collision data collection process recommendations. For data to be available for analysis at the national level, it will be necessary to retain SSO agencies and the FTA and its NTD. Redundancies in the way data move from organization to organization should, however, be avoided in a restructured national reporting system. The project makes three recommendations for the development of a national LRT colli- sion data reporting system: 1. A standardized electronic reporting form should be developed and made available to all local LRT agencies. 2. All collision reports should include relevant traffic exposure measures. Exposure meas- ures are required to put each collision in its local context and to allow for meaningful aggregation and comparison at the national level. 3. Details of location geometry and all safety and traffic control devices should be collected. These details are required to support a high quality and meaningful national analysis of the safety effects of LRT safety treatments. Summary of Recommendations Following is a summary of the nine major recommendations put forward in this report. Details about each recommendation are provided in the chapter cited in brackets. 1. Use proxies to assess the effectiveness of specific safety treatments without needing to wait for a significant number of collisions [Chapter 4]. 2. Follow four general treatment strategies [Chapter 4]: a) Give responsibility to the operators, b) Increase motorist, pedestrian, and cyclist awareness through active, appropriate information, c) Education, and d) Separate LRT space from the space occupied by other modes. 3. Continue to add to the LRT Catalog of Safety Treatments provided in Appendix A [Chapter 5]. 4. Use a standard LRT risk analysis methodology that addresses all of the elements listed in the LRT Alignment Risk Assessment Checklist provided in Chapter 6 [Chapter 6]. 5. Develop and implement a standardized, comprehensive, electronic LRT incident report- ing form [Chapter 7]. 6. Provide a reporting form structure that can be easily transferred into a searchable elec- tronic database [Chapter 7]. 7. Ensure that LRT collision reports include fields for storing geometric details and traffic exposure measures that apply to the specific incident site [Chapter 7]. 8. To support national-level analysis of LRT safety, an LRT crossing database should be cre- ated. This database should include for each alignment location (crossing or segment): details of geometry, control devices, and traffic exposure [Chapter 7]. 9. In the standardized electronic LRT incident reporting form, consider using fields that provide the most valuable information for researchers, designers, and operators, as listed in Chapter 7 [Chapter 7].