Integrating Observational and Traffic Simulation Models for Priority Ranking of Unsafe Intersections

Abstract

Observational models based on reported crash history are the most common measures for identifying unsafe sites for priority intervention. Observational models are good for predicting higher-severity crashes but ignore higher-risk vehicle interactions (e.g., near misses) that failed to result in crashes that are reported in historical data. Proponents of microscopic simulation models argue that failure to recognize these higher-risk interactions can significantly understate the safety problem at a given site and lead to misallocation of scarce treatment funds. This paper takes the position that a complete understanding of the safety problem at a given site can emerge only if both crash potential and traffic conflicts are taken into account. A priority-ranking model that integrates estimates from observational crash prediction models into an analysis of traffic conflicts is presented. Traffic conflicts were based on simulated vehicle interactions and deceleration requirements for different traffic scenarios. The suitability of the approach for priority ranking of sites was assessed with six ranking approaches: crash frequency, empirical Bayes, potential for safety improvement, conflict frequency, conflict rate (sum and cross product of traffic volume), and integrated model. Priority ranking was evaluated with five test criteria: site consistency, method consistency, rank difference, total rank score, and sensitivity and specificity. These models were applied to a sample of 58 signalized intersections from Toronto, Ontario, Canada, for the period from 1999 to 2006. The integrated model was found to yield better results for the five evaluation criteria; this result suggests that the proposed approach has promise.

Using Microsimulation to Evaluate Safety and Operational Implications of Newer Roundabout Layouts for European Road Networks

“…It can be seen that the target roundabout has the highest absolute and normalized values (per 1000 entering vehicles). Finally, the calibration parameters for "Lane change" and "Lateral" were taken from the research conducted by Shahdah et al [16]. Figure 4 shows the comparison of absolute and normalized operational performance measures for different roundabout layouts for entry flows at 80% of capacity.…”

Empirical Bayesian Before-and-After Analysis of Treatments: Simulated Traffic Conflict Approach

“…On the other hand, the 0.50s TTC threshold ref lects a much higher risk wherein a driver only have 0.50s or less in which to take appropriate action to avoid a crash. These TTC thresholds have been used by other researchers (Shahdah et al, 2014;Cunto, 2008;Archer, 2005;etc.). The number of traffic conflicts at each site was calculated as an average from the 50 simulation runs.…”

“…A conflict is observed when two or more vehicles approach each other, such that collision can take place if the vehicles trajectories unchanged (Amundsen and Hyden, 1977). The use of traffic conflicts in traffic safety analysis is based on that traffic conflicts can address some of the statistical issues associated with the rare random nature of crashes and reporting problems, especially for low severity crashes (Shahdah, 2014;Shahdah et al, 2014). In addition, the transportation failure mechanism, using traffic conflicts, is considered from a broader causal perspective than from observational analysis (Shahdah et al, 2014;Brown, 1994;Sayed et al, 1994;Van der Horst, 1990).…”

“…Microscopic traffic simulation platforms (e.g. VISSIM (PTV, 2012), SUMO (Lopez, 2018), AIMSUN (TSS, 2019) and PAR AMICS (Quadstone, 2019)) can be used in the estimation of vehicle interactions and hence traffic conflicts (Gettman and Head, 2003;Cunto and Saccomanno, 2008;Shahdah et al, 2014;Shahdah et al, 2015;etc. ) The study introduced in this paper suggests that simulated traffic conflicts can be used in place of traffic volumes to enhance the performance of EB before-and-after analysis for estimating treatment effects.…”

Traffic conflicts analyses for 2+1 road sections

“…VISSIM) are post-processed. The application of microsimulated conflicts is considered as one of the methods based on surrogate measures which allows to estimate the number of conflicts for different road and traffic configurations [8][9][10][11][12]. Such an approach is much more popular in the analysis of intersections and, although simulation modeling for road sections with additional lanes is described in literature, these limited studies mainly discuss the modeling in the context of the evaluation of traffic performance [13,14].…”