Application of traffic microsimulation for evaluating safety performance of urban signalized intersections. (November 2015)

• Record Type: Journal Article
• Authors: Shahdah, Usama; Saccomanno, Frank; Persaud, Bhagwant
• Is Part Of: Transportation research. Volume 60(2015)
• Page Start: 96
• Page End: 104
• Publication Date: 2015-11
• Subjects: Crash modification factors -- Safety performance -- Traffic microsimulation -- VISSIM -- Conflicts -- Surrogate measures; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2015.06.010
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: Provides a statistical relationship between simulated conflicts and observed crashes. The paper addresses a major limitation of microsimulation. It provides an alternative to crash based evaluations for estimating safety effects. It provides a unique application of microsimulation results from VISSIM. It assesses the effect of key microsimulation factors on the estimated safety effects. Abstract: The primary objective of this paper is to provide a statistical relationship between traffic conflicts estimated from microsimulation and observed crashes in order to evaluate safety performance, in particular the effect of countermeasures. A secondary objective is to assess the effect of conflict risk tolerance and number of simulation runs on the estimates of countermeasure effects so obtained. Conflicts were simulated for a sample of signalized intersections from Toronto, Canada, using VISSIM microscopic traffic simulation and several crash–conflict relationships were obtained. A separate sample of treated intersections from Toronto was used to compare countermeasure effects from the integrated crash–conflict expression to a conventional, but rigorous crash-based Empirical Bayes before-and-after analysis that was already done, with the results published, for the same sites and treatment. The countermeasure considered for this investigation involved changing the left turn signal operation for the treated intersection sample from permissive to protected-permissive. The results support the view that countermeasure effects can be estimated reliably from conflicts derived from microsimulation, and more so when a suitable number of simulation runs and conflict tolerance thresholds are used in the crash–conflict relationship.