Predicting Likelihood of Hit-and-Run Crashes Using Real-Time Loop Detector Data and Hierarchical Bayesian Binary Logit Model with Random Effects

Fairly extensive research studies have been dedicated to identifying the influential factors of hit-and-run (HR) crashes. Most of them utilized the typical Maximum Likelihood Estimation Binary Logit models, and none of them have employed the real-time traffic data. To fill this gap, the study focused on predicting the likelihood of HR crashes, as well as the general ones, based on the hierarchical Bayesian models with random effects within a sequential Logit structure. Two-year crash and real time loop detector data were collected from one freeway segment in Southern California. The results indicated similar significant contributing factors to general crashes as in the previous research. As for the HR crashes, the factors of upstream vehicle speed, roadway segment length, and weekend were found to be positively correlated with the HR crash risk. k-fold cross validation technique resulting in ROC (Receiver Operating Characteristic) curves exhibited the satisfactory prediction accuracy of the developed models in the study. Overall, the research suggested that the real-time traffic data seems to be a promising area of interest for understanding the conditions of HR crashes.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee ANB20 Standing Committee on Safety Data, Analysis and Evaluation.
  • Corporate Authors:

    Transportation Research Board

    500 Fifth Street, NW
    Washington, DC  United States  20001
  • Authors:
    • Xie, Meiquan
    • Cheng, Wen
    • Gill, Gurdiljot Singh
    • Falahati, Roya
    • Jia, Xudong
    • Choi, Simon
  • Conference:
  • Date: 2017


  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: 16p
  • Monograph Title: TRB 96th Annual Meeting Compendium of Papers

Subject/Index Terms

Filing Info

  • Accession Number: 01628185
  • Record Type: Publication
  • Report/Paper Numbers: 17-06376
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 8 2016 12:35PM