Integration of immune features into a belief-desire-intention model for multi-agent control of public transportation systems

Abstract

There is a growing need to develop monitoring and control systems to maintain the performance and the quality of service of Public Transportation Systems (PTS) at acceptable levels, especially in case of traffic disturbances, such as accidents or congestion. Despite the use of Multi-Agent Systems (MAS) to control PTS, many existing systems still rely on centralized control architectures, and do not offer generic agent behavior models. Many Belief-Desire-Intention (BDI) models were developed as generic agent decision-making processes, but existing developments still lack detailed descriptions of models instantiation and implementation. This article introduces a new framework for the implementation of the Belief-Desire-Intention (BDI) model for the development of an agent based decision support system for the control of public transportation systems. The suggested framework uses a set concepts and mechanisms inspired from biological immunity. Through a simulation case study, we have presented an example of implementation of the suggested BDI framework.