Preferred time headway of drivers on two-lane highways with heterogeneous traffic

Abstract

This paper presents an investigation on the adequacy of one of the major performance indicators for two-lane highways, the percent-time-spent-following (PTSF) under changing driver’s behavior in choosing headways. Field study was conducted on a two-lane highway in India that exhibits heterogeneity in its traffic composition and a wide range of flow levels was covered while collecting data. The appropriate headway distributions were selected based on goodness-of-fit test. Exponential or Erlang distributions exhibit their aptness in describing headways up to a volume to capacity ratio of 0.3 and gamma and exponential distributions were compatible for headway data corresponding to a volume to capacity ratio of 0.4 and above. Around 40% drivers were observed to adopt headways 2 s or less at heavy flow when a car-following situation prevails; these headways are considered unsafe in several instances. Further, speed–headway relationships were established to assess a headway threshold beyond which vehicles could be considered free. Empirical investigation indicates a value of about 6 s under such traffic unlike 3 s as considered in PTSF estimation. Another major implication on PTSF is a few drivers move in a platoon by choice even if they have a gap of more than 6 s and passing opportunities.

Keywords:

• Time headway
• two-lane highways
• heterogeneous traffic
• platoon
• percent-time-spent-following
• Kolmogorov–Smirnov (K–S) test