Skip to main navigation Skip to search Skip to main content

Investigating traffic safety reckoning hyperbolic driving following behavior using trajectory data

  • Delft University of Technology
  • Sardar Vallabhbhai National Institute of Technology

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

Using vehicle trajectory datasets developed over a study section for three traffic flow levels, a rectangular hyperbolic relation between time-to-collision and relative speeds in vehicle-following behavior were observed. A new methodology for estimating the probable rear-end collisions in the given traffic stream is developed based on this relation. The vehicle-following behavior is examined in terms of the hysteresis phenomenon concerning the distance gap (DG) and relative speed (RS). Further, based on the follower's attention toward towards the leader vehicle, a novel surrogate safety measure, called Instantaneous Heeding Time (IHT), was conceptualized. This measure represents the time gap available based on the positions of the leader and follower vehicles. After exploring vehicle-following behavior, IHT, DG, and RS were used to estimate the rear-end collision probability. The applicability of the proposed methodology is tested using different thresholds (IHT, DG, and RS) and applied to the study section at the three traffic flow levels.

Original languageEnglish
Article number128129
JournalPhysica A: Statistical Mechanics and its Applications
Volume606
DOIs
StatePublished - 15 Nov 2022

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 3 - Good Health and Well-being
    SDG 3 Good Health and Well-being
  2. SDG 11 - Sustainable Cities and Communities
    SDG 11 Sustainable Cities and Communities

Keywords

  • Hyperbolic vehicle-following
  • Hysteresis
  • Instantaneous heeding time
  • Rear-end collisions

Fingerprint

Dive into the research topics of 'Investigating traffic safety reckoning hyperbolic driving following behavior using trajectory data'. Together they form a unique fingerprint.

Cite this