Applying Safety Data and Analysis to Performance-based Transportation Planning
- 1. National Regional Transportation Conference Presented to: National Association of Development Organizations (NADO) June 14, 2016
- 2. • Applying Safety Data and Analysis to Performance Based Transportation Planning • http://safety.fhwa.dot.gov/tsp/fhwasa15089/ • http://safety.fhwa.dot.gov/tsp/fhwasa15089/data_anl.pdf New Guidebook
- 3. • Address severe crashes on public roads in nonmetropolitan areas • Low cost countermeasures to address high risk roadway features at multiple locations • Help the state meet fatality and serious injury targets for all public roads • Consider safety within the context of all transportation projects to make the future system safer Importance of Safety Planning
- 5. • Institutional considerations necessary to initiate and sustain data collection efforts • Types of safety data and how they can inform transportation planning • Methods to obtain data to initiate data analysis Data Collection Chapter
- 6. • Who manages the statewide crash database? • What data are available and in what format? • What data do I need? • How do I obtain access to data? Data Collection Approach: Questions to Ask/Answer
- 7. Common Types of Traffic Safety Data
- 8. • For each crash that meets a minimum injury or property damage reporting threshold, a large set of information is collected and entered into a State database • Where, why, what, and how fatalities and serious injuries are occurring Crash Data Model Minimum Uniform Crash Criteria (MMUCC) Crash, Vehicles or Units Involved, People Involved
- 9. • Set of information that describes the physical attributes and conditions of the street network • Associate crashes with road features to tell more of the crash story – Risk factors, performance of different facility types Roadway Data Model Inventory of Roadway Elements (MIRE) Functional class, surface type, shoulders, medians
- 10. • All States are required to collect and submit traffic count data to understand traffic volume • Ability to understand safety risks based on the amount of vehicle travel – crash rates Traffic Volume FHWA’s Highway Performance Monitoring System (HPMS)
- 11. • Citation/Adjudication – arrest and conviction records • Injury – emergency department data, discharge data, trauma registry • Vehicle – vehicle technologies, types of vehicles, other vehicles (CMV) Other Data
- 12. • Raw Data – not aggregated or significantly modified from original state • Generated Reports – prepared summary report based on a template • Web Interfaces/Tools – access to crash data and other datasets through Web interfaces • Custom Inquiries – summary of information based on a specific request Obtaining Safety Data
- 13. DATA ANALYSIS
- 14. • Basic safety analysis categories and tools • Applications to transportation planning • Examples Safety Analysis Chapter
- 15. Analysis Category Analysis Question Benchmarking How many fatalities and serious injuries are occurring in my area? How does this compare to other areas or my State? Identify Crash Trends and Contributing Factors Who is involved in crashes? When are the crashes occurring? What are the major contributing factors to crashes? Identify and Evaluate Focus Crash Types What are the most common crash types? What are the most common contributing factors? What are the characteristics of the over representation? Network Screening—Identify Sites for Safety Improvement What locations (intersections or segments) show the most potential for safety improvements? Systemic Analysis—Identify Safety Risk Factors What are the common characteristics of locations with crashes? What are the countermeasures to address these characteristics? How should we prioritize system wide implementation? Corridor and Intersection Planning Safety Analysis What are the safety effects of alternative roadway or intersection cross sections? Analysis Categories
- 16. Analysis Categories - Appendix B
- 17. • Understanding the scope and scale of safety concerns in the planning area Benchmarking
- 18. • Identify major crash trends and contributing factors to develop countermeasures Crash Trends and Contributing Factors
- 19. • Understand crashes by type, severity, over-representation, or geography Crash Types
- 20. • Identify sites (intersections, segments) that would benefit from safety improvements – Identify Sites (crash frequency, crash rate, EPDO) – Prioritize Sites – Field Investigations and Recommended Countermeasures Network Screening
- 21. Network Screening Final Rank Intersection Traffic Control Frequency Rank EPDO Rank Crash Rate Rank Composite Score 1 U.S. 60 at RadanovichBoulevard at GlobeFood Mart access road Signal 4 2 5 11 2 SR 87/BeelineHighway atLonghorn RoadatU.S. 260 Signal 1 9 2 12 3 SR 260 at Manzanita DriveatGraniteDells Road Signal 4 12 7 23 4 SR 260 at Valley Road at Highline Drive Stop Sign 11 7 6 24 5 Broad Street at Oak Street Stop Sign 7 16 1 24 6 SR 87/BeelineHighway atBonita Street Signal 2 10 14 26 7 SR 188/Apache Trail atU.S. 60atRussellRoad Signal 7 8 15 30 8 U.S. 60/Ash StreetatHill Street Signal 6 15 11 32 9 U.S. 60 at Escudilla Drive at Main Street Signal 3 26 4 33 10 SR 347/John Wayne Parkway at PapagoRoad Stop Sign 14 3 24 41
- 23. • Reducing crash frequency and severity through implementation of low-cost safety improvements to address high-risk roadway features correlated with specific severe crash types – Risk Factors: 2-lane roads with specific curve radii – Crash Types: Roadway departures – Identify all locations where this combination exists – Prioritize and implement low cost countermeasures at all locations Systemic Analysis
- 24. • Calculates expected number and severity of crashes at sites with similar geometric and operational characteristics – existing conditions, future conditions, or roadway design alternatives Highway Safety Manual – Predictive Analysis
- 26. How to use data and analysis to: • Develop goals, objectives, performance measures • Identify and prioritize safety programs and projects Applying Safety Data and Analysis to Inform Decisionmaking Chapter
- 27. • Trends – Demonstrate a need for a general safety goal • Contributing factors – Safety specific goals Using Safety Analysis to Set Goals
- 28. • Goal – Reduce severe crashes on the transportation system • Objective – Identify solutions to address the overrepresentation of pedestrian and motorcycle crashes Using Safety Analysis to Set Objectives
- 29. Using Safety Data to Set/Track Performance Measures
- 30. Using Safety Analysis for Project Prioritization/Programming
- 31. • Training on utilizing safety data and analysis in rural safety planning process • Input on content for: – Data Collection – Data Analysis – Application to the Planning Process Next Steps – Breakout Groups
Editor's Notes
- #2: This guidebook is not one that gives instructions on how to collect data and conduct data analysis. Rather, it shows how data and analysis can be applied to performance based planning processes.
- #3: The primary audience is planners. However, we have discovered that others are interested in the guidebook as it touches upon setting goals, objectives, and targets. Basically, the guidebook help answers questions like: What types of data are available and how to obtain it What safety analysis tools are available and how they work How results can be used for decisionmaking
- #7: State DOT, Office of Highway Safety, Department of Motor Vehicles, University, Other Crash, roadway, vehicles mile traveled, available data formats What am I trying to accomplish Knowing who to ask
- #9: Crash data is the foundation of transportation safety planning. The specific crash data elements that are collected vary by State, but generally are based on the guidelines outlined in the Model Minimum Uniform Crash Criteria (MMUCC). Officers who respond to the scene of a crash collect information on the crash itself (e.g., location, number of vehicles, type of crash); the vehicles or ‘units’ involved (e.g., model and year, estimated speed); and the people involved (e.g., driver, occupant, pedestrian, age, sex, impairment status). One of the most important data fields in the crash report is the injury level of each person involved in the crash. The KABCO system is a standardized way of coding injury severity using the following injury codes: K—fatality; A—suspected serious injury; B—suspected minor injury; C—possible injury; and O—no apparent injury.
- #10: MIRE includes a detailed list of data elements that are recommended for collection by transportation agencies for the specific purpose of evaluating safety. Some examples of information collected in text box – over 100 elements Associating crashes – for example, it may show you that a number of crashes are occurring in areas with no shoulders or that a large number of crashes are occurring on two-lane versus a four lane rural roads. Challenging to collect roadway data, especially on local routes
- #11: Traffic volume data are essential to understanding the safety risk based on the amount of travel. For instance, if planners are conducting a safety review of intersections, they may discover that intersections with the same number of crashes can have very different risk profiles depending on the level of traffic.
- #13: Raw - Although working with raw data affords the greatest flexibility, it can be time-consuming and resource-intensive to work with raw data files. Staff with data management and analysis skills are needed to effectively leverage raw data. Generated - Reports generated by State agencies are likely to provide an overall understanding of crash trends at the given reporting level. For example, they may report on fatality and serious injury trends, provide an indication of the most prevalent crash types or contributing factors in an area, or identify high-crash locations. For planners who do not have the time or resources to analyze raw data, these reports provide a basic amount of information to initiate transportation safety planning. Web Interfaces - This is a convenient format that makes it easy for planners to access data, while also reducing the need for data owners to respond to data requests. It is common for some level of analysis capabilities to be incorporated into these data access tools. Users may have the ability to filter data based on criteria such as year, type of crash, injury severity, geographic location, or other factors. From there, they can download the raw data or generated reports based on the selected criteria. Custom - request data on an as-needed basis
- #15: This Chapter provides an overview of the basic safety analysis categories, questions answered in each category, how this can be included in transportation planning, and analysis methods to answer these questions.
- #16: This table summarizes the analysis categories and what questions they can answer. A more detailed version of the Table is included in Appendix B of the Guidebook. That table also include methods and tools necessary to complete the analysis and data needs. This particular chapter provides an overview of the basic safety analysis categories, including examples for how to complete the analysis.
- #17: An excerpt of the detailed table in Appendix B mentioned in the previous slide. It includes information regarding analysis categories, questions asked, available tools, and types of data. Next slide will show examples of these two categories: bench marking and network screening.
- #18: Fundamentally, benchmarking like this makes it possible to monitor safety performance within the area, as well as assess performance of the system relative to other comparable areas and/or the State. At the highest level, these performance measures quantify the number of people killed or seriously injured over time in the area. Monitored annually, this will provide the agency with information about overall effectiveness and safety planning and programming in the area and will help transportation agencies determine the appropriate safety focus areas moving forward.
- #19: Descriptive statistics count and compare the number (i.e., frequency), type, severity of crashes, and/or contributing factors that have occurred historically. Typically, this is done for the most recent three- to five-year period at a site, along a corridor or in an area. As described in the bullets above, there are many possible contributing factors and categories to count. As such, it is advisable to start with a small number of categories to keep the process manageable. The figure on the right shows one example of looking at crash trends by driver age – this would help identify educational and/or enforcement programs to the target group.
- #20: Planner would review and summarize crash data to understand such topics as: What is the manner of collision? Rear-end, angle, roadway departure, head-on? Was a pedestrian or bicycle involved? What is the crash severity? Which crash types are most severe and where are they occurring? What crash types are overrepresented? What is the geographic distribution of the focus crash type? What are common roadway features? The scatterplot (figure 5.8) illustrates a simple method for evaluating the relative occurrence of various crash types in terms of frequency and severity. The x-axis represents the crash type frequency rank, while the y-axis indicates the crash type severity rank (percentage of crashes that result in a severe injury). In this example, the single-vehicle (RD) crash type stands out as being highly ranked in both frequency and severity. At the other extreme, backing and other crashes are ranked in the bottom in both frequency and severity and, therefore, are relatively unimportant.
- #21: In a network screening analysis specific locations (also known as sites) are identified and prioritizes for safety improvements. It identifies locations that are experiencing more crashes than would be expected for comparable sites and would benefit from safety improvements.
- #22: Count of crashes 2009-2013 Crash rate – count of crashes divided by traffic volume Equivalent Property Damage Only (EPDO) - Crash frequency weighted by a crash severity cost index
- #27: Once safety data are collected and analyzed, the results can be used for decisionmaking. This Chapter provides information on how to develop data-driven safety goals, objectives, and performance measures. More importantly, it shows how to use data to identify and prioritize safety programs and projects.
- #28: Increase safety for all users by reducing transportation-related fatalities and serious injuries Improve safety by reducing the prevalence of infrastructure, impaired driving, occupant protection, and young driver-related fatalities. Other examples: Crash Types - Improve multimodal safety by reducing crashes on principal arterials throughout the State. Network Screening - Improve safety by reducing fatalities and serious injuries at intersections.
- #29: Regardless of the approach used to establish transportation safety goals, objectives are necessary to define how the goals will be met. Ultimately, objectives should provide enough specificity to help planners identify programs and projects that will meet the goals of the planning document. From here, you could develop strategies, such as Identify low-cost pedestrian safety improvements for local roads; Promote pedestrian public education campaigns, including the shared use of roadways; and Facilitate coordination among jurisdictions and business to address missing pedestrian links.
- #30: Performance measures are a useful tool to understand programmatic progress towards transportation goals and objectives
- #31: Tying together goals, objectives, performance measures and targets, are transportation and safety projects and programs. With limited resources, programs and projects should be identified, evaluated, and prioritized based on whether or not they address (or have the likelihood to address) the goals in the plan and the extent to which they contribute to meeting performance targets. Over time, it is possible to see the effect programs and projects have on the goals of the plan.