| Nowadays, the number of people who use motor vehicles to work grows rapidly. Traffic problems become more and more serious. In this paper, an analysis of commuters' parking choice behavior is described. It can help government to guide the distribution of traffic flow in road network by the commuters'parking management, and then ease the traffic congestion.In this article, it firstly summarizes the early studies of parking behavior at home and abroad. According to previous research results, a RP and SP survey of the drivers'parking lots choice is conducted by the parking behavior questionnaire around DiWang building in Shenzhen city. According to the questionnaire data, the commuters' social economic characteristics and regulation of parking behavior are analyzed.Based on preliminary data research, the multinomial logit model is introduced by the disaggregate theory. There are three parking choice ML models: for all commuters, for the commuters whose parking fees are paid by government and for the commuters whose parking fees are paid by themselves. The parameters of all models are analyzed and tested by SPSS. Then the fixed term of utility function is composed of the travel time of parking road, parking fees, walking time and alternative specific dummy variables.According to the parking choice models calibration, an optimization parking assignment model can be formulated. It is a logit-based stochastic user equilibrium model. The Logit model is used to describe the parking choice behavior. The Wardrop user equilibrium principle is adopted to describe the route choice behavior. It can help people to understand the relationship between dynamic and static traffic more deeply. This model is solved by partial linear algorithm. The calculation results show that the model and the algorithm are effective in practical case. Finally, two simple applications of the assignment model are presented. |
