Multi-objective supervisory flow control based on fuzzy interval arithmetic: Application for scheduling of manufacturing systems.
Résumé
Complex production systems can produce more than one part type under multiple and possibly conflicting objectives. This paper considers the design of the multiple objective real-time scheduling problem of a multiple-part-type production system. Based on fuzzy control theory and fuzzy arithmetic intervals, distributed and supervised continuous-flow control architecture has been proposed. The objective is to balance the production process by adjusting the continuous production rates of the machines on the basis of the average behaviour. The supervisory control aims to maintain the overall performance within acceptable limits. In the proposed approach, the problem of a real-time scheduling of jobs is considered at the shop-floor level. In this case, the actual dispatching times are determined from the continuous production rates through a discretisation procedure. To deal with conflicts between jobs at a shared machine, a decision for the actual part to be processed is taken using some criterions which represent a measure of the job's priority. A case study demonstrates the efficiency of the proposed control approach.