ABSTRACT
Optimizing operations in semiconductor manufacturing facilities is challenging. The production line can be modeled as an NP-hard constrained flexible job-shop scheduling problem, intractable with mathematical optimization due to its scale. Therefore, decision-making in factories is dominated by handcrafted heuristics. Though machine learning-based approaches proved efficient in solving such problems, their applications are limited due to the lack of trust in the underlying black-box models, and issues with scalability for larger instances. This work presents a genetic programming-based method to generate explainable, improved dispatching heuristics. Our method outputs a set of human-readable dispatching strategies, verifiable by scheduling experts before deployment. In case of minor changes in the environment or the optimization objectives, the continued evolution of the candidate solutions is possible without starting the training process from scratch. The introduced method is evaluated on a simulator executing real-world scale instances. The resulting heuristics improve the key performance indicators of the generated schedules. Furthermore, the generated dispatchers are easy to integrate into existing industrial systems. These favorable properties make the method applicable to various large-scale, dynamic, practical scheduling scenarios, where adaptions to different environments go along with modest human effort limited to the design of a fitness function.
- Mickaël Bureau, Stéphane Dauzère-Pérès, and Yazid Mati. 2006. SCHEDULING CHALLENGES AND APPROACHES IN SEMICONDUCTOR MANUFACTURING. IFAC Proceedings Volumes 39, 3 (2006), 739--744. 12th IFAC Symposium on Information Control Problems in Manufacturing. Google ScholarCross Ref
- Xiaokun Chang, Ming Dong, and Dong Yang. 2013. Multi-objective real-time dispatching for integrated delivery in a Fab using GA based simulation optimization. Journal of Manufacturing Systems 32, 4 (2013), 741--751. Google ScholarCross Ref
- Tsung-Che Chiang. 2013. Survey: Enhancing Rule-Based Scheduling in Wafer Fabrication Facilities by Evolutionary Algorithms: Review and Opportunity. Comput. Ind. Eng. 64, 1 (jan 2013), 524--535. Google ScholarDigital Library
- Domagoj Jakobović and Leo Budin. 2006. Dynamic Scheduling with Genetic Programming. In Genetic Programming, Pierre Collet, Marco Tomassini, Marc Ebner, Steven Gustafson, and Anikó Ekárt (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 73--84.Google Scholar
- Chihyun Jung and Tae-Eog Lee. 2012. An Efficient Mixed Integer Programming Model Based on Timed Petri Nets for Diverse Complex Cluster Tool Scheduling Problems. IEEE Transactions on Semiconductor Manufacturing 25, 2 (2012), 186--199. Google ScholarCross Ref
- Chihyun Jung, Detlef Pabst, Myoungsoo Ham, Marcel Stehli, and Marcel Rothe. 2014. An Effective Problem Decomposition Method for Scheduling of Diffusion Processes Based on Mixed Integer Linear Programming. IEEE Transactions on Semiconductor Manufacturing 27, 3 (2014), 357--363. Google ScholarCross Ref
- Andreas Klemmt, Jan Lange, Gerald Weigert, Frank Lehmann, and Jens Seyfert. 2010. A multistage mathematical programming based scheduling approach for the photolithography area in semiconductor manufacturing. In Proceedings of the 2010 Winter Simulation Conference. 2474--2485. Google ScholarCross Ref
- Denny Kopp, Michael Hassoun, Adar Kalir, and Lars Mönch. 2020. SMT2020---A Semiconductor Manufacturing Testbed. IEEE Transactions on Semiconductor Manufacturing 33, 4 (2020), 522--531. Google ScholarCross Ref
- B. Kovács, P. Tassel, R. Ali, M. El-Kholany, M. Gebser, and G. Seidel. 2022. A Customizable Simulator for Artificial Intelligence Research to Schedule Semiconductor Fabs. In Proceedings of the Thirty-third Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC'22). IEEE Computer Society Press, 106--111.Google Scholar
- John R Koza, David Andre, Forrest H Bennett III, and Martin A Keane. 1996. Use of automatically defined functions and architecture-altering operations in automated circuit synthesis with genetic programming. In Proceedings of the First Annual Conference on Genetic Programming. Stanford University MIT Press, Cambridge, MA, 132--140.Google ScholarCross Ref
- Andreas Kuhnle, Jan-Philipp Kaiser, Felix Theiß, Nicole Stricker, and Gisela Lanza. 2021. Designing an adaptive production control system using reinforcement learning. Journal of Intelligent Manufacturing 32, 3 (2021), 855--876. Google ScholarDigital Library
- Won-Jun Lee, Byung-Hee Kim, Keyhoon Ko, and Hayong Shin. 2019. Simulation Based Multi-Objective Fab Scheduling by Using Reinforcement Learning. In Winter Simulation Conference Proceedings. 2236--2247. Google ScholarCross Ref
- Kazuo Miyashita. 2000. Job-Shop Scheduling with Genetic Programming. In Proceedings of the 2nd Annual Conference on Genetic and Evolutionary Computation (Las Vegas, Nevada) (GECCO'00). Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 505--512.Google ScholarDigital Library
- Jatoth Mohan, Krishnanand Lanka, and A. Neelakanteswara Rao. 2019. A Review of Dynamic Job Shop Scheduling Techniques. Procedia Manufacturing 30 (2019), 34--39. Digital Manufacturing Transforming Industry Towards Sustainable Growth. Google ScholarCross Ref
- Lars Mönch, John W. Fowler, Stéphane Dauzère-Pérès, Scott J. Mason, and Oliver Rose. 2011. A survey of problems, solution techniques, and future challenges in scheduling semiconductor manufacturing operations. Journal of Scheduling 14, 6 (01 Dec 2011), 583--599. Google ScholarCross Ref
- Su Nguyen, Yi Mei, and Mengjie Zhang. 2017. Genetic programming for production scheduling: a survey with a unified framework. Complex & Intelligent Systems 3, 1 (01 Mar 2017), 41--66. Google ScholarCross Ref
- In-Beom Park, Jaeseok Huh, Joongkyun Kim, and Jonghun Park. 2020. A Reinforcement Learning Approach to Robust Scheduling of Semiconductor Manufacturing Facilities. IEEE Transactions on Automation Science and Engineering 17, 3 (2020), 1420--1431. Google ScholarCross Ref
- Christoph Pickardt, Jürgen Branke, Torsten Hildebrandt, Jens Heger, and Bernd Scholz-Reiter. 2010. Generating dispatching rules for semiconductor manufacturing to minimize weighted tardiness. In Proceedings of the 2010 Winter Simulation Conference. 2504--2515. Google ScholarCross Ref
- Joc Cing Tay and Nhu Binh Ho. 2008. Evolving dispatching rules using genetic programming for solving multi-objective flexible job-shop problems. Computers & Industrial Engineering 54, 3 (2008), 453--473. Google ScholarDigital Library
- Jeffrey Voas, Nir Kshetri, and Joanna F. DeFranco. 2021. Scarcity and Global Insecurity: The Semiconductor Shortage. IT Professional 23, 5 (2021), 78--82. Google ScholarDigital Library
- Bernd Waschneck, Thomas Altenmüller, Thomas Bauernhansl, and Andreas Kyek. 2016. Production Scheduling in Complex Job Shops from an Industry 4.0 Perspective: A Review and Challenges in the Semiconductor Industry. In SAMI@iKNOW.Google Scholar
- Annie S. Wu and Robert K. Lindsay. 1995. Empirical Studies of the Genetic Algorithm with Noncoding Segments. Evolutionary Computation 3, 2 (1995), 121--147. Google ScholarDigital Library
- Cheng-Hung Wu, Fang-Yi Zhou, Chi-Kang Tsai, Cheng-Juei Yu, and Stéphane Dauzère-Pérès. 2020. A deep learning approach for the dynamic dispatching of unreliable machines in re-entrant production systems. International Journal of Production Research 58, 9 (2020), 2822--2840. arXiv:https://doi.org/10.1080/00207543.2020.1727041 Google ScholarCross Ref
- Fangfang Zhang, Yi Mei, Su Nguyen, and Mengjie Zhang. 2021. Evolving Scheduling Heuristics via Genetic Programming With Feature Selection in Dynamic Flexible Job-Shop Scheduling. IEEE Transactions on Cybernetics 51, 4 (2021), 1797--1811. Google ScholarCross Ref
Index Terms
- Optimizing Dispatching Strategies for Semiconductor Manufacturing Facilities with Genetic Programming
Recommendations
Adaptive scheduling on unrelated machines with genetic programming
Graphical abstractDisplay Omitted HighlightsA scheduling procedure consisting of a meta-algorithm and priority function is described.A genetic programming approach used to generate priority functions for scheduling on unrelated machines.A comparative ...
Feature Selection in Evolving Job Shop Dispatching Rules with Genetic Programming
GECCO '16: Proceedings of the Genetic and Evolutionary Computation Conference 2016Genetic Programming (GP) has been successfully used to automatically design dispatching rules in job shop scheduling. The goal of GP is to evolve a priority function that will be used to order the waiting jobs at each decision point, and decide the next ...
Evolving reusable operation-based due-date assignment models for job shop scheduling with genetic programming
EuroGP'12: Proceedings of the 15th European conference on Genetic ProgrammingDue-date assignment plays an important role in scheduling systems and strongly influences the delivery performance of job shops. Because of the stochastic and dynamic features of job shops, the development of general due-date assignment models (DDAMs) ...
Comments