Genetic learning of fuzzy rules applied to sequencing problem of FMS

Several techniques have been used for the determination of a good sequencing of parts that are stored in queues waiting for be processed. These techniques aim to improve the use of factory's resources, to increase the productivity, to decrease the lead time, to maintain the delivery date, etc. In this context, this work presents an approach that use fuzzy system to set a more suitable parts sequencing to be processed at the machines. The fuzzy rule base of this system is generated from data using a genetic algorithm. In order to test and validate the proposed approach, a shop floor was tested using the fuzzy system obtained.