A localised fuzzy-neural fluctuation smoothing rule for job scheduling in a wafer fab

Abstract

This paper presents a localised fuzzy-neural fluctuation smoothing rule to improve the performance of scheduling jobs in a wafer fabrication factory (wafer fab). The rule is modified from the tailored non-linear fluctuation smoothing (TNFS) rule with some innovative treatments. First, the remaining cycle time of a job is forecasted with an evolving fuzzy-neural approach in order to improve the accuracy. Second, in the original TNFS rule, the adjustable factor is static, and in this rule it becomes dynamic. Third, the adjustable factor in the new rule depends on the jobs gathering before the same machine, and the TNFS rule becomes localised. To assess the effectiveness of the proposed methodology, production simulation is also applied in this study. According to the experimental results, the proposed methodology is better than some existing approaches in reducing the average cycle time and cycle time standard deviation.