Experimental study on the departure time choice behavior during the morning rush hours under different road capacity conditions

This paper presents an experimental study on the departure time choice behavior during the morning rush hours under different road capacity conditions. Experimental data are analyzed from the aspects of the equilibrium state of a traffic system and the choice behavior of subjects. The experimental results showed that the user equilibrium is easy to achieve in the medium-capacity scenario; however, it is difficult in the low- and high-capacity scenario. This implies that the user equilibrium cannot predict the aggregate behavior well when the bottleneck capacity is too low or too high. A reinforcement learning model is constructed to reproduce experimental results and uncover subjects' learning mechanism. Simulation results are in good agreement with the experimental results. The results presented in this study could provide the theoretical support for developing measures for transportation management and control during the morning rush hours.