A Monotone Approximate Dynamic Programming Approach for the Stochastic Scheduling, Allocation, and Inventory Replenishment Problem: Applications to Drone and Electric Vehicle Battery Swap Stations: Applications to Drone and Electric Vehicle Battery Swap Stations
Saved in:
| Title: | A Monotone Approximate Dynamic Programming Approach for the Stochastic Scheduling, Allocation, and Inventory Replenishment Problem: Applications to Drone and Electric Vehicle Battery Swap Stations: Applications to Drone and Electric Vehicle Battery Swap Stations |
|---|---|
| Authors: | Amin Asadi, Sarah Nurre Pinkley |
| Source: | Transportation Science. 56:1085-1110 |
| Publication Status: | Preprint |
| Publisher Information: | Institute for Operations Research and the Management Sciences (INFORMS), 2022. |
| Publication Year: | 2022 |
| Subject Terms: | FOS: Computer and information sciences, Computer Science - Machine Learning, Battery degradation, math.OC, Battery swap station, Computer Science - Artificial Intelligence, Approximate dynamic programming, cs.LG, Probability (math.PR), 0211 other engineering and technologies, Regression-based initialization, 02 engineering and technology, cs.AI, Electric vehicles and drones, math.PR, Monotone policy and value function, 7. Clean energy, Markov decision processes, Machine Learning (cs.LG), Artificial Intelligence (cs.AI), Optimization and Control (math.OC), FOS: Mathematics, 22/1 OA procedure, Mathematics - Optimization and Control, Mathematics - Probability |
| Description: | There is a growing interest in using electric vehicles (EVs) and drones for many applications. However, battery-oriented issues, including range anxiety and battery degradation, impede adoption. Battery swap stations are one alternative to reduce these concerns that allow the swap of depleted for full batteries in minutes. We consider the problem of deriving actions at a battery swap station when explicitly considering the uncertain arrival of swap demand, battery degradation, and replacement. We model the operations at a battery swap station using a finite horizon Markov decision process model for the stochastic scheduling, allocation, and inventory replenishment problem (SAIRP), which determines when and how many batteries are charged, discharged, and replaced over time. We present theoretical proofs for the monotonicity of the value function and monotone structure of an optimal policy for special SAIRP cases. Because of the curses of dimensionality, we develop a new monotone approximate dynamic programming (ADP) method, which intelligently initializes a value function approximation using regression. In computational tests, we demonstrate the superior performance of the new regression-based monotone ADP method compared with exact methods and other monotone ADP methods. Furthermore, with the tests, we deduce policy insights for drone swap stations. |
| Document Type: | Article |
| Language: | English |
| ISSN: | 1526-5447 0041-1655 |
| DOI: | 10.1287/trsc.2021.1108 |
| DOI: | 10.48550/arxiv.2105.07026 |
| Access URL: | http://arxiv.org/pdf/2105.07026 http://arxiv.org/abs/2105.07026 |
| Rights: | CC BY |
| Accession Number: | edsair.doi.dedup.....e028ce7e92a74ac1d4f49378e742b975 |
| Database: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstract: | There is a growing interest in using electric vehicles (EVs) and drones for many applications. However, battery-oriented issues, including range anxiety and battery degradation, impede adoption. Battery swap stations are one alternative to reduce these concerns that allow the swap of depleted for full batteries in minutes. We consider the problem of deriving actions at a battery swap station when explicitly considering the uncertain arrival of swap demand, battery degradation, and replacement. We model the operations at a battery swap station using a finite horizon Markov decision process model for the stochastic scheduling, allocation, and inventory replenishment problem (SAIRP), which determines when and how many batteries are charged, discharged, and replaced over time. We present theoretical proofs for the monotonicity of the value function and monotone structure of an optimal policy for special SAIRP cases. Because of the curses of dimensionality, we develop a new monotone approximate dynamic programming (ADP) method, which intelligently initializes a value function approximation using regression. In computational tests, we demonstrate the superior performance of the new regression-based monotone ADP method compared with exact methods and other monotone ADP methods. Furthermore, with the tests, we deduce policy insights for drone swap stations. |
|---|---|
| ISSN: | 15265447 00411655 |
| DOI: | 10.1287/trsc.2021.1108 |