摘要

In this paper, we incorporate activity-based trip chaining behavior into the network equilibrium analysis. An integrated model which combines Beckman-type congestion link terms and entropy-type logit demand terms is proposed to describe the traveler behavior. The convexity and equivalency conditions of the model are discussed. Based on the integrated model, a bi-level model is designed to maximize the social welfare by appropriate parking fee. Also, an expanded network is developed to eliminate the non-additivity of the utilities of activities and travelling in the original network. Then, the Simulated Annealing (SA) method is used to solve the proposed bi-level model. Numerical examples are presented to examine the model's availability and effects of the parking fee scheme on traveler behavior and social welfare. Results show that the model is effective in describing the trip chaining behavior in the network.