Fuzzy production control with limited resources and response delay

This paper attempts to study the possibility of using fuzzy control in production-inventory systems that contain variable processes with fixed capacity. The average processing rate that builds up the inventory is equal to the average delivery rate that consumes the inventory so that the system needs no-control action on the basis of averages. However, the inherent randomness in both consumption and replenishment of inventory gives rise to random variations in inventory level that needs to be controlled. Attempt is thus made to develop a fuzzy control system (FCS) to control the effects of randomness in such production systems. Effects of various parameters of the FCS on various performance measures are investigated so that a FCS may be developed for specific situations. It is shown how limited resources on production capacity and the delay due to processing time frustrate the efforts of designing an effective FCS although some benefits can still be achieved. The question of how frequently the fuzzy inference engine (FIE) should observe the system and process the inputs is also studied. Finally it is shown how a little extra production capacity makes the FCS very effective. The investigations provide an understanding and methodology that can be used to design FCS for any situation and, although we have considered two products, it can easily be extended to more than two products.