Multimodal Transportation Network Design Using Physarum Polycephalum-Inspired Multi-agent Computation Methods

Abstract

In this paper, a new approach towards P. Polycephalum inspired computational efforts is proposed, with specific application to the problem of Multimodal transportation network design for planned cities of the future. Working with a multi-agent model of the Physarum Polycephalum, parallels are drawn between agent properties and mode characteristics, and agents are allowed to dynamically change from one mode to another. A mechanism to compare the performance of resultant multimodal networks against single mode networks involving the same component modes is demonstrated. The observations point to the potential applicability of the new approach in city planning and design.