The emergence of Industry 4.0 allows for new approaches to solve industrial problems such as the Job Shop Scheduling Problem. It has been demonstrated that Multi-Agent Reinforcement Learning approaches are highly promising to handle complex scheduling scenarios. In this work we propose a user friendly Multi-Agent Reinforcement Learning tool, more appealing for industry. It allows the users to interact with the learning algorithms in such a way that all the constraints in the production floor are carefully included and the objectives can be adapted to real world scenarios. The user can either keep the best schedule obtained by a Q-Learning algorithm or adjust it by fixing some operations in order to meet certain constraints, then the tool will optimize the modified solution respecting the user preferences using two possible alternatives. These alternatives are validated using OR-Library benchmarks, the experiments show that the modified Q-Learning algorithm is able to obtain the best results.
Jiménez, YM, Palacio, JC & Nowé, A 2020, Multi-Agent Reinforcement Learning Tool for Job Shop Scheduling Problems. in B Dorronsoro, P Ruiz, JCDL Torre, D Urda & E-G Talbi (eds), Optimization and Learning - 3rd International Conference, OLA 2020, Proceedings. vol. 1173, Communications in Computer and Information Science, vol. 1173 CCIS, Springer, pp. 3-12, The International Conference in Optimization and Learning, Spain, 17/02/20. https://doi.org/10.1007/978-3-030-41913-41
Jiménez, Y. M., Palacio, J. C., & Nowé, A. (2020). Multi-Agent Reinforcement Learning Tool for Job Shop Scheduling Problems. In B. Dorronsoro, P. Ruiz, J. C. D. L. Torre, D. Urda, & E.-G. Talbi (Eds.), Optimization and Learning - 3rd International Conference, OLA 2020, Proceedings (Vol. 1173, pp. 3-12). (Communications in Computer and Information Science; Vol. 1173 CCIS). Springer. https://doi.org/10.1007/978-3-030-41913-41
@inproceedings{9ac6c01aca7e46c5ae7f1eb29930f85b,
title = "Multi-Agent Reinforcement Learning Tool for Job Shop Scheduling Problems",
abstract = "The emergence of Industry 4.0 allows for new approaches to solve industrial problems such as the Job Shop Scheduling Problem. It has been demonstrated that Multi-Agent Reinforcement Learning approaches are highly promising to handle complex scheduling scenarios. In this work we propose a user friendly Multi-Agent Reinforcement Learning tool, more appealing for industry. It allows the users to interact with the learning algorithms in such a way that all the constraints in the production floor are carefully included and the objectives can be adapted to real world scenarios. The user can either keep the best schedule obtained by a Q-Learning algorithm or adjust it by fixing some operations in order to meet certain constraints, then the tool will optimize the modified solution respecting the user preferences using two possible alternatives. These alternatives are validated using OR-Library benchmarks, the experiments show that the modified Q-Learning algorithm is able to obtain the best results.",
author = "Jim{\'e}nez, {Yailen Mart{\'i}nez} and Palacio, {Jessica Coto} and Ann Now{\'e}",
note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2020.; The International Conference in Optimization and Learning, OLA2020 ; Conference date: 17-02-2020 Through 19-02-2020",
year = "2020",
doi = "10.1007/978-3-030-41913-41",
language = "English",
isbn = "9783030419127",
volume = "1173",
series = "Communications in Computer and Information Science",
publisher = "Springer",
pages = "3--12",
editor = "Bernab{\'e} Dorronsoro and Patricia Ruiz and Torre, {Juan Carlos de la} and Daniel Urda and El-Ghazali Talbi",
booktitle = "Optimization and Learning - 3rd International Conference, OLA 2020, Proceedings",
url = "https://ola2020.sciencesconf.org/",
}