This paper presents an optimized-rule-fusion and certain-rule-first approach to improve the performance of scheduling jobs in a wafer fabrication factory. As a nonlinear fusion of four traditional dispatching rules, the new rule is aimed at the simultaneous optimization of the average cycle time, cycle time standard deviation, the maximum lateness, and the number of tardy job at the same time, which has rarely been discussed in the past. In addition, we show that there is a contradiction among the four objectives, and, establish a certain-rule-first procedure to resolve the contradiction. A more effective fuzzy-neural approach is also applied to estimate the remaining cycle time of a job. Finally, the values of the adjustable parameters in the nonlinear fusion are optimized using response surface method (RSM) instead of being chosen subjectively as in the previous studies. Some theoretical properties of the new rules are also proven. To assess the effectiveness of the proposed methodology, production simulation is also applied in this study. According to the results of a simulation study, the proposed methodology is better than several existing approaches in improving the four objectives at the same time.
|Number of pages||20|
|Journal||International Journal of Innovative Computing, Information and Control|
|State||Published - 17 Jul 2013|
- Multi-objective scheduling
- Nonlinear fusion
- Response surface method
- Wafer fabrication