Domain of impact for agents collaborating in a baggage handling system

Abstract

Recognize the frustration of waiting in slow starting queues? Acceleration ramps and reaction times make traffic queues frustrating to many people every day. Similar problems arise in many production and material handling systems. In this paper we present research activities conducted on a baggage handling system (BHS) of a large airport hub in Asia, where we have applied an intelligent multi-agent based approach to control the flow of bags on the BHS. By exchanging a centralized control system with an agent-based solution, local queues can be avoided or minimized, which increase the overall performance of the BHS. Through an established community of highly collaborating and coordinating agents, each agent can, based on its relative placement in the topology of the BHS, decide whether it is appropriate to route more bags through this node relative to the overall system load. The agent-based approach not only improves robustness of the system, and utilize the entire BHS in a more convenient and dynamical way, it also include strategies for maximizing capacity of the system. We present results from ongoing work of developing suitable and proficient algorithms and agent collaboration schemes to increase the performance of the BHS. In this paper we pay special attention to the impact of the relative physical displacement of the agents in the system. © 2007 International Federation for Information Processing.