Agent-based decision-making process in airport ground handling management

Abstract

Recent research on highly distributed control methods has produced a series of new philosophies based on negotiation, which bring together the process engineering with computer science. Among these control philosophies, the ones based on Multi-agent Systems (MAS) have become especially relevant to address complex tasks and to support distributed decision making in asset management, manufacturing, and logistics. However, these MAS models have the drawback of an excessive dependence on up-to-date field information. In this work, a theoretical and experimental MAS, called MAS-DUO, is presented to test new strategies for managing handling operations supported by feedback coming from radio frequency identification (RFID) systems. These strategies have been based on a new distributed organization model to enforce the idea of division between physical elements and information and communication technologies (ICT) in the product scheduling control. This division in two platforms simplifies the design, the development, and the validation of the MAS, allowing an abstraction and preserving the independency between platforms. The communication between both platforms is based on sharing the parameters of the Markov reward function. This function is mainly made up of the field information coming from the RFID readers incorporated as the internal beliefs of the agent. The proposed MAS have been deployed on the Ciudad Real Central Airport in Spain in order to dimension the ground handling resources.