Decentralized scheduling of baggage handling using multi-agent technologies

Agile and lean manufacturing are the trendy buzz-words among manufacturers today. Lowvolume high-variety product series have radically changed the conditions for manufacturing systems. To handle the new challenges and continuously optimize production flow, new requirements for flexible manufacturing systems have been introduced. Moving away from dedicated hardware and mass production additionally stress the disturbances of a dynamic environment. Whereas dedicated hardware are optimized system-wide and not expected to perform unless it is fully operational, flexible manufacturing systems should deal with these circumstances in the line of product variety, because the production usually allow the system to continue under restricted conditions. It seems evident that not only are these systems harder to build in hardware, as they should be able to handle different products, but controlling them in an optimal manner is far from being a trivial development task. In this chapter we will present novel approaches and strategies for developing a multi-agent based solution to control a baggage handling system of a larger airport hub. The baggage-handling system (BHS) is similar in setup, complexity, and operation to many manufacturing systems. Items enter the system at input facilities, the procurement of the system, undergo some processing at various stations during its way to the output facilities – shipping in classic production terminology. The research activities being described are conducted on a real case study of a BHS in Asia, and the project is part of a larger research project, called DECIDE, which seeks to prove and evaluate the feasibility of using multi-agent based control in large manufacturing systems. The project has been supported by the Ministry of Science, Technology, and Innovation in Denmark. The chapter is structured as follows. We start by giving a detailed introduction to the evolution of manufacturing systems towards the flexible setup suitable for agent-based control. Next we provide an introduction to the FIPA specifications used to generalize agent interoperability and design. Then we give a general introduction to the baggage handling problem, and briefly describe the case of the Denver International Airport, which among airport managers still may be a warning to experiment with intelligent baggage handling systems. O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m