A robust optimal scheduling system based on multi-performance driving for complex manufacturing systems.

Area of Science:

• Operations Research
• Manufacturing Systems Engineering
• Artificial Intelligence

Background:

• Complex manufacturing systems face challenges in balancing dynamic scheduling, equipment maintenance, and overall robustness.
• Existing scheduling schemes often struggle to adapt to mixed uncertainties and integrate maintenance tasks effectively.

Purpose of the Study:

• To propose a robust optimal scheduling method for collaborative optimization of dynamic scheduling and preventive equipment maintenance.
• To enhance the robustness of scheduling schemes in complex manufacturing environments.
• To improve overall production performance under uncertainty.

Main Methods:

• Mapping equipment maintenance to the process level and designing composite dispatching rules with weight parameters.
• Analyzing performance using two models based on the IWOA-MLP algorithm.
• Implementing a feedback mechanism for adaptive closed-loop optimizations.

Main Results:

• The proposed system effectively handles mixed uncertainties in manufacturing.
• Demonstrated significant improvements in production performances.
• Maintained highly robust scheduling measures in simulation experiments.
• Validated on a semiconductor manufacturing enterprise simulation platform.

Conclusions:

• The developed robust optimal scheduling system offers a viable solution for complex manufacturing challenges.
• The integration of dynamic scheduling, maintenance, and adaptive feedback mechanisms leads to enhanced efficiency and robustness.
• The method shows practical applicability in real-world manufacturing scenarios, as evidenced by semiconductor industry simulations.