Fuzzy-neural-network-based fluctuation smoothing rule for reducing the cycle times of jobs with various priorities in a wafer fabrication plant: A simulation study

Tin-Chih Chen*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

23 Scopus citations

Abstract

This paper presents a fuzzy-neural-network-based fluctuation smoothing rule to further improve the performance of scheduling jobs with various priorities in a wafer fabrication plant. The fuzzy system is modified from the well-known fluctuation smoothing policy for a mean cycle time (FSMCT) rule with three innovative treatments. First, the remaining cycle time of a job is estimated by applying an existing fuzzy-neural-network-based approach to improve the estimation accuracy. Second, the components of the FSMCT rule are normalized to balance their importance. Finally, the division operator is applied instead of the traditional subtraction operator in order to magnify the difference in the slack and to enhance the responsiveness of the FSMCT rule. To evaluate the effectiveness of the proposed methodology, production simulation is applied to generate some test data. According to the experimental results, the proposed methodology outperforms six existing approaches in the reduction of the average cycle times. In addition, the new rule is shown to be a Pareto optimal solution for scheduling jobs in a semiconductor manufacturing plant.

Original languageEnglish
Pages (from-to)1033-1043
Number of pages11
JournalProceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture
Volume223
Issue number8
DOIs
StatePublished - 1 Aug 2009

Keywords

  • FSMCT
  • Fuzzy
  • Neural
  • Scheduling
  • Simulation
  • Tailored
  • Wafer fabrication

Fingerprint Dive into the research topics of 'Fuzzy-neural-network-based fluctuation smoothing rule for reducing the cycle times of jobs with various priorities in a wafer fabrication plant: A simulation study'. Together they form a unique fingerprint.

Cite this