@article{ref1,
title="NET-RAT: Non-equilibrium traffic model based on risk allostasis theory",
journal="Transportation research part A: policy and practice",
year="2023",
author="Mohammadian, Saeed and Zheng, Zuduo and Haque, Mazharul and Bhaskar, Ashish",
volume="174",
number="",
pages="e103731-e103731",
abstract="Empirical studies of vehicle trajectories have shown that psychological theories of driver behaviour can shed light on car-following processes and the associated empirical traffic phenomena. Numerous continuum models have been derived from car-following relations in order to model macroscopic traffic flow dynamics through collective intercations between car-following processes. However, the existing continuum models cannot capture the psychological processes underlying drivers' car-following in accordance to behavioural thoeries, and thus, have little implications for investigating empirical traffic phenomena in relation to human psychological factors. This paper develops a novel continuum model (Non-Equilibrium Traffic Model based on Risk Allostasis Theory, i.e., NET-RAT) from a car-following model, extended in this work, by incorporating drivers' behavioural adaptions in relation to perceived risk. We first extend the full-velocity difference car-following model (FVDM) to incorporate drivers' pereception of risk and its impacts on drivers' adaptation time to frontal stimuli using car-following information (e.g., speed, spacing, etc) and corresponding safety surroage measures. Risk allostasis theory is used to model the impacts of perceived risk on drivers' stimulus-response behavioural adaptation. We then derive NET-RAT by continuum approximations of the extended FVDM and the corresponding behavioural components. Theoretical investigations show that NET-RAT has desirable analytical properties regarding macroscopic traffic flow dynamics, and that such properties can also be explained meaningfully from a behavioural perspective. Furthermore, we investigate NET-RAT's performance for real-world traffic by using data from the German A5 autobahn in order to study the impacts of drivers' risk perception on the complex real-world traffic phenomena (e.g., wide scattering, traffic instabilities, hysteresis, etc.). The investigations show that NET-RAT can meaningfully capture such complex phenomena by linking them to the interplay between drivers' dynamic adaption process in relation to their perceived car-following risk.<p /> <p>Language: en</p>",
language="en",
issn="0965-8564",
doi="10.1016/j.tra.2023.103731",
url="http://dx.doi.org/10.1016/j.tra.2023.103731"
}