Identifying High-Crash Intersections

Birchler Arroyo Associates, Inc. is assisting the City of Novi with a City-wide study of high-crash intersections. The initial phase is to collect 5 years of traffic and crash data for about 60 major intersections and determine which intersections qualify as "high-crash."

When traffic crash data is typically collected at the municipal level, communities often focus on the top-ten intersections based on crash rate. Unlike evaluations based simply on crash frequency, those based on crash rates – crashes per million entering vehicles – reflect the true risk of having a crash.

While studying crash rates within the boundary of a municipality is helpful, drilling down to understand whether or not a crash rate is significant or not can provide more in-depth insight into high-crash intersections. By comparing local data to a large sample of physically similar intersections serving similar traffic volumes – rather than a smaller sample of more diverse intersections within the City – it can be more reliably determined that a location’s crash history should truly be a matter of concern. Following a methodology included in the Southeast Michigan Council of Government's (SEMCOG) Traffic Safety Manual, crash rates are compared to a Critical Crash Rate to guard against drawing unwarranted conclusions based on a location’s crash rate being higher than average due to random occurrences. The Critical Crash Rate is based on a statistical confidence interval developed using data collected from similar intersections in the southeast Michigan region. Only where the crash rate exceeds the Critical Crash Rate is the intersection reliably classified as “High-Crash.”

In addition, crash severity is also evaluated and is expressed as a casualty ratio – the proportion of all crashes involving at least one fatality or non-fatal injury. To guard against drawing unwarranted conclusions based on a location’s casualty ratio being higher than average due to random occurrences, the casualty ratio is compared to a Critical Casualty Ratio. Only where the casualty ratio exceeds the Critical Casualty Ratio is the intersection reliably classified as “High-Crash-Severity.” As with the Critical Crash rate, the Critical Casualty ratio comes from a statistical confidence interval developed using data collected from similar intersections in the region.

Besides identifying high-crash intersections that may warrant further study to identify crash mitigation measures, this project has resulted in updating the crash data base with the most current traffic count data. This has several potential uses and benefits in addition to providing the most reliable traffic crash rate at each of the intersections evaluated.

It is expected that this study will be concluded in the next 60 days.