PLANNING SNAPSHOT 7:

Transcription

1 PLANNING SNAPSHOT 7: CRASH DATA JANUARY 2016 Funded through the NCHRP 8-36 Research Series, these snapshots are designed to tell you a little about the current state of a specific planning practice of interest today. Crash Data Applications for Planning To better understand how crash data is collected, analyzed, and applied for planning, transportation planners and crash data managers were surveyed. This survey was distributed on behalf of SCOP, AMPO, and NARC. 33 state agencies and 27 regional organizations responded providing the insights and information shared here. State DOT MPO/Region CRASH DATA WIDELY USED FOR PLANNING Beyond traditional applications, safety data is increasingly informing planning and decision-making How is safety and crash data typically used for planning purposes? State DOT MPO/Region Long-range planning Funding prioritization Corridor- or project-level analyses Performance reporting and tracking 78.3% 75.0% 78.3% 75.0% 75.0% 78.3% 87.0% 91.7% Grant and programming decision-making 66.7% 82.6% Systems analysis 29.2% 78.3% Other 4.2% 13.0% Not typically used for planning purposes 0.0% 8.7% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Other applications of safety data include: project scoping, design and operational analysis, diagnostic assessment, safety specific studies, sketch planning, and community engagement activities NCHRP 08-36, Task 120 Crash Data for Planning 1

2 PLANNING SNAPSHOT 7: CRASH DATA ANALYZING CRASH DATA YIELDS INSIGHTS Safety data is commonly used to identify trends and pinpoint needs, but also increasingly integrated with agency-wide management and information systems 22 What analyses are commonly performed and what can the data tell you? State DOT MPO/Region High-level basic crash and fatality reporting Hot spots or roadway location identification Contributing factors or risk analysis Detailed crash characteristics and conditions Is crash data integrated with other transportation data or information systems? No, systems are not well integrated Asset management systems 14% 16% We lack a common roadway link length and ID. Publicly available reports Systems are partially integrated 21% 23% We are a small organization (2 FTE), so staffing is an issue. Highway performance monitoring systems or transportation data 32% Performance management systems 32% Geographic information systems 61% Safety data is available in asset management, performance management, HPMS and transportation, GIS and publicly available reports, but it is not automatic or efficient, it takes time and effort. 2 NCHRP 08-36, Task 120 Crash Data for Planning

3 BICYCLE AND PEDESTRIAN CRASH DATA WIDELY SOUGHT States and MPOs agree, bicycle and pedestrian volumes would be helpful for planning and policy What data is currently not available that would be helpful in policy and planning? Bicycle volumes 17 Pedestrian volumes Intersection data Comprehensive local crash data Traffic volumes 2 7 State DOT MPO/Region Roadway characteristics 7 Other 5 2 IN-DEPTH LOOK AT BICYCLE AND PEDESTRIAN CRASH DATA NEEDS Despite challenges, many agencies collect bicycle and pedestrian crash data involving a motor vehicle Does your agency collect bicycle and pedestrian crash data? No (7) Yes (37) If no, what are some challenges to collecting bicycle/ pedestrian data Lack of resources, or low priority Low-severity crashes often not reported or underreported Difficult to collect data on crashes that do not involve a motor vehicle If yes, what types of bicycle/pedestrian data are collected? 29 Specific crash characteristics of bicycle/ pedestrian incidents Only reports of incidents that involve a motor vehicle, but pursuing more advanced reporting 6 Report of bicycle/ pedestrian incidents not involving a motor vehicle NCHRP 08-36, Task 120 Crash Data for Planning 3

4 PLANNING SNAPSHOT 7: CRASH DATA CRASH DATA PROCESSES ARE STREAMLINED AND EFFICIENT, BUT COORDINATION CHALLENGES PERSIST Does your agency have difficulty providing processed data to safety analysts in time for analysis? Yes, frequently Yes, occasionally Rarely or not at all 10% 33% 57% Typical timeline of data collection to analysis From crash event to database entry typically takes 1-3 months Data from the crash database is most often analyzed on an ongoing basis and typically occurs within 1 year What agency is primarily responsible for collecting and long-term storage of crash reports and data? What agency is primarily responsible for analyzing, utilizing, and distributing crash data for planning purposes? 45% 40% 75% 25% 15% Dept. of Public Safety/State Police Dept. of Transportation Dept. of Motor Vehicles University or Contractor Dept. of Public Health 4 NCHRP 08-36, Task 120 Crash Data for Planning

5 DATA CHALLENGES COMMON, EVEN WITH INCREASED INTEREST AND APPLICATION OF SAFETY DATA What challenges or obstacles has your agency encountered in working with, analyzing, or sharing crash data? Incomplete state or local data 60% Inaccurate data Legal or liability concerns 43% 45% Incompatible datasets 35% Agency barriers 28% CRASH DATA LOCATION AND QUALITY ASSURANCE, multiple responses allowed States use both automated and manual processes to conduct QA/QC on crash data How does your state generate location data for crash reports? Crash locations recorded at scene via GPS 27% Crash locations recorded at scene based on mileage marker or other visible landmarks 23% Crash locations automatically recorded at scene via software 20% Crash locations identified from reports during data entry Some combination of above methods 13% 17% How is locational accuracy and quality control conducted? Person at agency cross-checks against police reports 58% Software-driven process at agency (e.g. ArcGIS) 29% Combination of automatic and manual checks 13% NCHRP 08-36, Task 120 Crash Data for Planning 5

6 PLANNING SNAPSHOT 7: CRASH DATA VARIETY OF TOOLS USED TO MANAGE CRASH DATA Many agencies use some combination of proprietary software and in-house tools to find storage and analysis options that meet their individual needs What databases do agencies use to store crash data? DO NOT ACCESS OTHER KNOW SQL ORACLE OTHER includes: DB2, ESRI File Geodatabase, agency dashboard PROS? 99Very fast and handles spatial data quickly 99From crash event to entry in database is less than three days 99More scalable and better performance custom queries are easy and the queries run quickly 99Allows flexibility in designing user interfaces 99Easy to use, familiar platform CONS? Expertise is lacking, state IT restricts many of the software s enhanced functions, and high performance is cost prohibitive Software takes time to learn and requires a lot of storage, need to know the intricacies of data relationships to ensure proper output data Program makes it difficult to maintain and organize data across many years What tools do agencies use to conduct analysis of crash data? 20 PROS? 99Tool is widely used and easy for most people 16 99Tool has a graphical interface 99Our in-house tool allows us to quickly query crash data and export to other programs 13 99Software is great for sharing data with the public CONS? Too few individuals proficient in the software Our in-house tool is based on very old technology and static network data 7 We offer a simplified tool for the public but there is no user manual and we often get asked questions Program requires too much data from many systems AASHTO WARE SAFETY ANALYST OTHER IN- HOUSE TOOL ArcGIS EXCEL OTHER includes: Tableau, R, Critical Analysis Reporting Environment 6 NCHRP 08-36, Task 120 Crash Data for Planning

7 MOST AGENCIES SHARE AND COMMUNICATE CRASH DATA All agencies release crash data by request, and over one-third of agencies also use online tools to disseminate crash data information Who has access to cleaned crash data? 30 respondents provide data to any member of the public or interested agency, by request 1 in 3 respondents provide summary crash data to the public online through mapping or data tools 1 in 5 respondents provide crash data to other state agencies and or local governments only Proportion of respondents How does your agency make crash data available to local and regional planning partners? 9% Examples of DOT and agency websites for communicating crash data are provided on the following page. Many state and regional agencies make safety data available and accessible in creative and innovative ways. 30% 32% 34% 43% Automated reporting from DOT Do not provide data Upon request only Via public website Static reporting from DOT, multiple responses allowed NCHRP 08-36, Task 120 Crash Data for Planning 7

8 PUBLIC DATA AND COMMUNICATIONS Ma ny s ta tes a nd re g io ns p ro vid e d eta ile d c ra s h re p o r t s a nd ma i nta i n c re a ti ve a nd a cc es s i ble c ra s h d a ta o nli ne. A fe w exa m ple s a re hig hlig hte d he re, mo re ca n b e fo u nd t h ro u g h N H T S A. Utah s Zero Fatalities website includes interactive statistics and data for key crash factors and characteristics. Arizona, Nevada, and Iowa maintain similar sites through the national zero fatalities initiative. See more at: The Louisiana Crash Data Reports website is a compilation of statistical data on a wide variety of topics linked to SHSP implementation. Crash data is maintained and visualized by LSU. See more at: MassDOT maintains an interactive map showing Top Crash Locations statewide; automated procedures were developed for processing, standardizing, matching, and aggregating the crash data by geographical location. See more at: For more information about this NCHRP effort and to view additional snapshots please visit Acknowledgment of Sponsorship This work was sponsored by the American Association of State Highway and Transportation Officials (AASHTO), in cooperation with the Federal Highway Administration, and conducted in the National Cooperative Highway Research Program (NCHRP) Project 08 36, Task 120, which is administered by the Transportation Research Board of the National Academies. Disclaimer The opinions and conclusions expressed or implied are those of the research agency that performed the research and are not necessarily those of the Transportation Research Board or its sponsors. The information contained in this document was taken directly from the submission of the author(s). This document is not a report of the Transportation Research Board or of the National Research Council. JANUARY 2016