FMS Scheduling Using Fuzzy Inference-guided Beam Search

Abstract

This paper proposes a beam search based FMS scheduling system where the evaluation function is adaptively selected throughout the scheduling period by utilizing a fuzzy inference based procedure. The proposed procedure is specially designed for dynamic production environments, where the conditions change frequently. In the proposed method, the performance of beam search with different evaluation functions is studied with varying production conditions. Environmental variables are then used to detect those environment changes, as input to the fuzzy inference system. The fuzzy inference will thereupon attribute preference levels to the evaluation functions, according to their adequacy to the current environment. The scheduling is made using partitioned combination of the evaluation functions. In order to verify the validity of the proposed method, an application to FMS scheduling is presented.