A Commodity Distribution Model for a Multi-Agent Freight System

Abstract

Modeling transportation systems at the microscopic level is currently gaining favor among researchers throughout the world. A multi-agent system, which is one method of microscopic modeling, is a dynamic framework dealing with the behavior of each individual actor in the transport system, and is especially favored for passenger transport models. Likewise, freight systems need a similar framework to better represent the interactions amongst freight agents, as large numbers of heterogeneous freight actors and other factors are involved. This paper proposes an extension of a multi-agent transport modeling system to cover the area of freight movement. The focus of this paper is at the stage of commodity distribution, in order to model the links between suppliers and receivers of commodities. The model is developed based on the principal of commodity movement through supply chains. The structure is a demand derived model, where receivers are the decision makers who choose their suppliers. The model is constructed based on the discrete choice method, considering the constraints of the amount of the commodity generated and the amount desired by each of the companies, travel impedance and attractive factors of the shippers. The model is developed and applied to the movement of urban freight in the Tokyo Metropolitan Area.