Fuzzy Decision Making for Agent Coordination in the Fractal Manufacturing System (FrMS)

Abstract

Fast adjustment of functionality and production capacity in manufacturing systems is one of the key requirements of survival in the rapidly changing markets and manufacturing environments. Fractal Manufacturing Systems (FrMS) is one of the promising system architectures that has such characteristics as flexibility, adaptability, self-similarity, and re-configurability to meet those requirements. A fractal in FrMS is a set of agents whose goal can be achieved through cooperation, coordination, and negotiation with other agents. Since each agent in the FrMS achieves its own goal autonomously, decision making in the FrMS can be seen as a distributed problem solving approach or distributed decision making approach. This paper presents an application of fuzzy decision making that can provide a coordination scheme to be used at the interactions among agents in FrMS. Fuzzy decision making provides an effective solution to the problem of forming and evaluating sub-goal/sub-task for each agent, and to the problem of handling negotiations of conflicting or competing goals among agents. Moreover, a global goal for a system cannot be divided into or represented as simple sets of sub-goals/sub-tasks for each component of the system, and this makes the agent coordination and the performance control of FrMS more complicated. Fuzzy-typed representation of the goal or performance for each agent enables the handling of those complex coordination problems in the distributed control system.