ABSTRACT
This paper proposes a deep reinforcement learning (DRL)-based framework for a novel variant of technician routing and scheduling problem. First, we model the problem as a Markov decision process (MDP). Then, we build the policy network with attention-based Graph Neural Network (GNN), autoregressive decoding, and sampling graph search technique. Finally, reinforcement learning (RL) is adopted in policy learning to overcome the difficulty of creating labelled datasets. Extensive computational results validate the efficacy of the framework and managerial insights are revealed for decision makers in real world practice.
- Amy Khalfay, Crispin Alan, and Crockett Keeley. 2017. A review of technician and task scheduling problems, datasets and solution approaches. 2017 intelligent systems conference (IntelliSys). IEEE, London, 288-296. https://doi.org/10.1109/IntelliSys.2017.8324306.Google ScholarCross Ref
- Matias S. Rasmussen, Tor Justesen, Anders Dohn, and Jesper Larsen. 2012. The home care crew scheduling problem: Preference-based visit clustering and temporal dependencies. European journal of operational research, Vol 219, 598-610. https://doi.org/10.1016/j.ejor.2011.10.048.Google ScholarCross Ref
- Victor Pillac, Guéret Christelle, and Andrés Medaglia. 2012. On the dynamic technician routing and scheduling problem. Proceedings of the Extended Abstracts of the 5th International Workshop on Freight Transportation and Logistics ODYSSEUS 2012. Mykonos, Greece, 509-512.Google Scholar
- Emilio Zamorano, and Raik Stolletz. 2017. Branch-and-price approaches for the multiperiod technician routing and scheduling problem. European Journal of Operational Research, Vol 257.1, 55-68. https://doi.org/10.1016/j.ejor.2016.06.058.Google ScholarCross Ref
- Yuchen Jiang, Xiang Li, Hao Luo, Shen Yin and Okyay Kaynak. 2022. Quo vadis artificial intelligence? Discover Artificial Intelligence, 2.1, 4. https://doi.org/10.1007/s44163-022-00022-8.Google ScholarCross Ref
- Mohammadreza Nazari, Afshin Oroojlooy, Lawrence V. Snyder, and Martin Takáč. 2018. Reinforcement learning for solving the vehicle routing problem. Advances in neural information processing systems, 9839–9849. https://doi.org/10.48550/arXiv.1802.04240.Google ScholarCross Ref
- Wouter Kool, Herke van Hoof, and Max Welling. 2019. Attention, Learn to Solve Routing Problems! International Conference on Learning Representations. https://doi.org/10.48550/arXiv.1803.08475.Google ScholarCross Ref
- Jingwen Li, Liang Xin, Zhiguang Cao, Andrew Lim, Wen Song, and Jie Zhang. 2021. Heterogeneous attentions for solving pickup and delivery problem via deep reinforcement learning. IEEE Transactions on Intelligent Transportation Systems, Vol 23.3, 2306-2315. https://doi.org/10.48550/arXiv.2110.02634Google ScholarCross Ref
- Liang Xin, Wen Song, Zhiguang Cao, and Jie Zhang. 2021. Multi-Decoder Attention Model with Embedding Glimpse for Solving Vehicle Routing Problems. Proceedings of the AAAI Conference on Artificial Intelligence, Vol 35, 12042-12049. https://doi.org/10.48550/arXiv.2012.10638.Google ScholarCross Ref
- Xi Chen, Barrett W. Thomas, and Mike Hewitt. 2016. The technician routing problem with experience-based service times. Omega, Vol 61, 49-61. https://doi.org/10.1016/j.omega.2015.07.006.Google ScholarCross Ref
- Marlin Ulmer, Maciek Nowak, Dirk Mattfeld, and Bogumił Kaminski. 2020. Binary driver-customer familiarity in service routing. European Journal of Operational Research, Vol 286, 477-493. https://doi.org/10.1016/j.ejor.2020.03.037.Google ScholarCross Ref
Index Terms
- Deep Reinforcement Learning for Solving Multi-period Routing Problem with Binary Driver-customer Familiarity
Recommendations
Deep Reinforcement Learning: From Q-Learning to Deep Q-Learning
Neural Information ProcessingAbstractAs the two hottest branches of machine learning, deep learning and reinforcement learning both play a vital role in the field of artificial intelligence. Combining deep learning with reinforcement learning, deep reinforcement learning is a method ...
Discrete-to-deep reinforcement learning methods
AbstractNeural networks are effective function approximators, but hard to train in the reinforcement learning (RL) context mainly because samples are correlated. In complex problems, a neural RL approach is often able to learn a better solution than ...
Multi-objective safe reinforcement learning: the relationship between multi-objective reinforcement learning and safe reinforcement learning
AbstractReinforcement learning (RL) is a learning method that learns actions based on trial and error. Recently, multi-objective reinforcement learning (MORL) and safe reinforcement learning (SafeRL) have been studied. The objective of conventional RL is ...
Comments