Production Scheduling of Semiconductor Wafer Fabrication Facilities Using Real-Time Combinatorial Dispatching Rule

Abstract

Fabrication of semiconductor wafers is a complicated and challenging process with huge system complexity and stochasticity in operations. To improve the above challenges and to enhance the system’s efficiency an effective scheduling technique is indispensable. In recent times Real-time scheduling with dynamic dispatching rules has been widely discussed. However, choosing the appropriate combination of dispatching rules in a dynamic environment is a challenge. In this study, we proposed a multi-objective non-dominated sorting genetic algorithm (MO-NSGA-II) approach for optimizing the dispatching rules by considering the multiple objectives as minimization of work in process, minimization of delay time, and minimization of makespan. To implement the proposed approach, the optimal parameters for the multi-objective evolutionary algorithm (MOEA) are selected based on the hierarchical combination method and by deploying the response surface methodology (RSM) the best combination of rules is generated. Further, a real-time simulated environment is created using Flexsim to check the significance of the proposed approach and the robustness of the generated combinatorial rules. Results stated that the proposed approach can improve the performance of a system to a greater extent.