Modelling a multi-modal logistic network with agents and dynamic graphs

Abstract

This paper presents a model of a logistic system. Our goal is to understand how such a system (with numerous stakeholders) behaves and evolves according to different constraints or scenarios. We adopted a complex system approach which leads us to propose an agent-based model coupled with dynamic graphs. It allows us to represent the properties, constraints and behaviours at a local level of a logistic system in order to reproduce the global behaviours thanks to the simulation in a dynamic context. The simulation (which uses data about the Seine axis) allows to test different scenarios in order to understand how local decisions impact the whole system. For example, this work presents the evolution of the system at the opening of the Seine-Nord Europe Canal. Indeed, this canal is a real major project for Europe, and has numerous economical stakes. So, we first describe the traffic evolution on the multi-modal transportation network (see figures 1 to 4). Then, we observe different other measures (evolution of costs, transportation mode share). Thanks to these analyses, we show that the Seine-Nord Europe Canal should promote the use of the river barges and reduce financial costs. In the same time, it could modify the respective shares of the northern European ports. Seine-Nord Europe Canal opened Day 125 (step 3000) Seine-Nord Europe Canal opened Day 125 (step 3000)Figure 4. River traffic (day 125) -Seine-Nord Europe Canal is closed.