Aligning Decision Support with Shop Floor Operations: A Proposal of Intelligent Product Based on BDI Physical Agents

MAS models have the drawback of an excessive dependence on up-to date field information and on complex interaction protocols. This work proposes a theoretical and experimental agent-based Decision Support System (DSS) architecture that is designed and developed to align shop floor operations, but including the Radio Frequency Identification (RFID) information feedback. Based on these automatic product feedbacks generated by the RFID visibility frameworks, the proposed Multi-Agent System (MAS) allows defining a competitive space where intelligent products negotiate by using their own knowledge and global/business constraints. Specifically, this product-driven MAS has been structured on a split organization model to enforce the idea of division between physical elements and “Information and Communication Technologies” (ICT). This division into two platforms simplifies the design, the development and the validation of the MAS in shop floor environments, providing a higher level of abstraction and preserving the independence between platforms. The proposed MAS framework, called MAS-DUO, has been tested in the ground handling operations at the Ciudad Real Central Airport and in a simulated logistics centre at the Autolog Labs-UCLM. This paper introduces the BDI physical agents of this framework as the core of this new approach, a new vision that mixes Beliefs-Desires-Intentions (BDI) reasoning, RFID and the Markov Decision Process (MDP).