Babak Mirbaha - Assistant Professor, Faculty of Engineering, Imam Khomeini International University, Qazvin, Iran
Ali Abdi Kordani - Assistant Professor, Faculty of Engineering, Imam Khomeini International University, Qazvin, Iran
Arsalan Salehikalam - PhD candidate, Imam Khomeini International University, Qazvin, Iran
Farzad Akbarinia - PhD candidate, Imam Khomeini International University, Qazvin, Iran

A Sudden speed drop in the leader vehicle of vehicle platoon results in propagating the deceleration wave from downstreamtowards the upstream flow. Points of wave propagation of the leader vehicle towards the follower vehicle identificationare done based on Newell’s theory in trajectory data. Deceleration wave propagates based on two parameters, time andspace, τ- δ. A follower driver performs different behavioural reactions that they result in deviating follower driver fromNewell’s trajectory. In this paper, follower driver behaviour was identified based on two theories. The asymmetricmicroscopic driving behaviour theory and traffic hysteresis were used during the deceleration and acceleration phases,respectively. The data trajectories were classified into different traffic phases. Driver’s parameters were identified at themicroscopic level. Since the follower driver had the nonlinear behaviour, artificial neural networks were developed. Theywere able to analysis and identify effective parameters of dependent variable between deceleration phases leading tocongestion phase, based on the behavioural patterns. Analysis results present effective parameters based on any behaviouralpatterns. Spacing difference of two phases, deceleration and congestion phases, was the most effective parameter of bothtwo behavioural patterns, under reaction – timid and over reaction – timid. Increasing the spacing difference of two phasesresults in decreasing (increasing) time based on under reaction – timid (over reaction – timid).

Stop–Go Traffic; Behavioural Patterns; Time between Two Phases; Deceleration Phase; Congestion Phase; Artificial NeuralNetworks