In order to study the evolution of day-to-day traffic network traffic under the toll road network at the departure time and the travel route, the travelers are divided into three categories according to the time value difference. According to historical experience, it updates the parameters related to the travel route, unifies the time value with the toll collection fee, and introduces the theory of prospect to describe the travel behavior of the traveler. The model of traffic transfer between departure time is established based on the maximum probability of punctual arrival. The objective of this method is to chase the maximum path foreground value and establish the traffic transfer model between each departure time. Finally, the model and algorithm are validated by examples. The results show that the traffic flow in the toll road segment and time segment is small, and the traffic flows to other road segments and time segments. Compared with homogeneous travelers, heterogeneous travelers tend to have more transitions and more moderate rates of overall traffic evolution.