The Virginia Department of Transportation (VDOT) is interested in refining work zone safety performance measures so that safety can be better assessed, monitored, and compared over time. VDOT currently uses the number of fatalities and the number of serious injuries in work zone traffic crashes as statewide performance measures. Although these are important direct safety metrics, they do not reflect differences in work zone exposure measures, such as the number of work zones. As a result, it is not always clear whether changes in work zone crashes are being caused by safety problems or changes in the amount of road work.
This study developed and improved statewide work zone safety performance measures for Virginia by combining information from two sources in VDOT: the VDOT crash database containing information in police crash reports and the Virginia Traffic Information Management System (VaTraffic) database containing information on incidents, work zones, weather conditions, etc. Count measures, such as the number of fatal work zone crashes, were calculated from the crash database. Exposure measures, such as work zone-hours, were calculated from the VaTraffic database. Combining these two types of measures resulted in rate measures such as number of fatal work zone crashes per 1,000 work zone-hours.
Unfortunately, existing databases still have a number of limitations that prohibit the calculation of ideal rate-based performance measures. Based on currently available data, four performance measures were found to be appropriate for monitoring and evaluating the statewide safety performance of work zones, and eight summary measures were identified in order to obtain further insights regarding statewide safety issues at work zones in Virginia. These measures were selected using the results of statistical analyses and input from a technical review panel consisting of work zone coordinators and traffic engineers.
An important finding of this study was that work zone safety assessments can change significantly depending on whether exposure measures are included. Based on count measures, work zones on non-interstate highways appear to have safety records equal to or better than those on interstate highways. However, when exposure measures were incorporated, work zones on interstate highways were shown to have better safety records. This study recommends that VDOT immediately begin using the performance measures identified in this study.