Work-Zone Safety Analysis: Evaluating Rear-End Crash Risk with Extreme Value Theory

Abstract

This research paper addresses the critical issue of work-zone safety, specifically focusing on the transition area where one or more lanes are reduced, leading to merging operations. Given the limited availability of reliable crash records, alternative methods are explored to evaluate safety in these zones. Surrogate safety measures (SSMs), such as time to collision (TTC), are employed to assess the potential risks. Videographic data is collected from two work-zone sites, located in Guwahati and Pune cities. The trajectories obtained from the data are analyzed to identify key parameters that influence the risk of rear-end crashes. To quantify merging-related crash risk, the study adopts the Gumbel distribution, which falls under the purview of extreme value theory (EVT). The EVT approach involves fitting the minimum TTC values, derived from the trajectory data, to the Gumbel distribution. This analysis allows for the estimation of crash risk associated with merging operations in work zones, providing valuable insights for implementing proactive safety measures. By focusing specifically on work-zone safety and exploring alternative evaluation methods, this research aims to overcome the challenges posed by the limited availability of crash records. The utilization of SSMs and the application of the Gumbel distribution within the framework of EVT offer comprehensive analysis of merging-related crash risk in work zones. Ultimately, the findings of this study aim to enhance safety and contribute to the implementation of effective measures to mitigate risks in work-zone areas, thereby ensuring the well-being of both workers and road users.