Loading [MathJax]/extensions/tex2jax.js

Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

  • Shimon Wonsak
  • Nils Henke
  • Mohammad Al-Rifai
  • Michael Nolting
  • Wolfgang Nejdl

Research Organisations

External Research Organisations

  • Volkswagen AG

Details

Original languageEnglish
Title of host publication32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Subtitle of host publicationACM SIGSPATIAL 2024
EditorsMario A. Nascimento, Li Xiong, Andreas Zufle, Yao-Yi Chiang, Ahmed Eldawy, Peer Kroger
Pages410-420
Number of pages11
ISBN (electronic)9798400711077
Publication statusPublished - 22 Nov 2024
Event32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024 - Atlanta, United States
Duration: 29 Oct 20241 Nov 2024

Abstract

Battery electric vehicles (BEV) offer an opportunity to decrease transportation and mobility emissions significantly. The availability of charging station networks and infrastructure is crucial for the proliferation of BEVs. While the expansion of the charging networks is still slow, optimal utilization of the existing infrastructure and dispatching of mobile charging stations can serve as a bypass while more charging stations are built. In this work, we propose a novel multi-agent reinforcement learning - AdapMCS - approach for optimizing the adaptive dispatching of mobile charging stations to maximize the number of served charging requests by a charging station operator while improving the customer experience. By combining graph neural networks with reinforcement learning our approach is able to adapt to dynamic spatio-temporal changes in the demand distribution, for example, during big events such as concerts or fairs. Furthermore, we conduct a thorough evaluation using a publicly available real-world dataset and simulation of dynamic demand distribution changes. The results show that our adaptive dispatching approach is able to deal with the demand shifts and achieve significant gains for both customers, in terms of reducing waiting and charging times, and operators, in terms of increasing their profit.

Keywords

    Graph Neural Networks, Mobile Charging Stations, Multi-Agent, Reinforcement Learning

ASJC Scopus subject areas

Cite this

Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning. / Wonsak, Shimon; Henke, Nils; Al-Rifai, Mohammad et al.
32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems: ACM SIGSPATIAL 2024. ed. / Mario A. Nascimento; Li Xiong; Andreas Zufle; Yao-Yi Chiang; Ahmed Eldawy; Peer Kroger. 2024. p. 410-420.

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Wonsak, S, Henke, N, Al-Rifai, M, Nolting, M & Nejdl, W 2024, Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning. in MA Nascimento, L Xiong, A Zufle, Y-Y Chiang, A Eldawy & P Kroger (eds), 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems: ACM SIGSPATIAL 2024. pp. 410-420, 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024, Atlanta, United States, 29 Oct 2024. https://doi.org/10.1145/3678717.3691262
Wonsak, S., Henke, N., Al-Rifai, M., Nolting, M., & Nejdl, W. (2024). Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning. In M. A. Nascimento, L. Xiong, A. Zufle, Y.-Y. Chiang, A. Eldawy, & P. Kroger (Eds.), 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems: ACM SIGSPATIAL 2024 (pp. 410-420) https://doi.org/10.1145/3678717.3691262
Wonsak S, Henke N, Al-Rifai M, Nolting M, Nejdl W. Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning. In Nascimento MA, Xiong L, Zufle A, Chiang YY, Eldawy A, Kroger P, editors, 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems: ACM SIGSPATIAL 2024. 2024. p. 410-420 doi: 10.1145/3678717.3691262
Wonsak, Shimon ; Henke, Nils ; Al-Rifai, Mohammad et al. / Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning. 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems: ACM SIGSPATIAL 2024. editor / Mario A. Nascimento ; Li Xiong ; Andreas Zufle ; Yao-Yi Chiang ; Ahmed Eldawy ; Peer Kroger. 2024. pp. 410-420
Download
@inproceedings{ed9df03326a04745b4457b78b221d7a5,
title = "Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning",
abstract = "Battery electric vehicles (BEV) offer an opportunity to decrease transportation and mobility emissions significantly. The availability of charging station networks and infrastructure is crucial for the proliferation of BEVs. While the expansion of the charging networks is still slow, optimal utilization of the existing infrastructure and dispatching of mobile charging stations can serve as a bypass while more charging stations are built. In this work, we propose a novel multi-agent reinforcement learning - AdapMCS - approach for optimizing the adaptive dispatching of mobile charging stations to maximize the number of served charging requests by a charging station operator while improving the customer experience. By combining graph neural networks with reinforcement learning our approach is able to adapt to dynamic spatio-temporal changes in the demand distribution, for example, during big events such as concerts or fairs. Furthermore, we conduct a thorough evaluation using a publicly available real-world dataset and simulation of dynamic demand distribution changes. The results show that our adaptive dispatching approach is able to deal with the demand shifts and achieve significant gains for both customers, in terms of reducing waiting and charging times, and operators, in terms of increasing their profit.",
keywords = "Graph Neural Networks, Mobile Charging Stations, Multi-Agent, Reinforcement Learning",
author = "Shimon Wonsak and Nils Henke and Mohammad Al-Rifai and Michael Nolting and Wolfgang Nejdl",
note = "Publisher Copyright: {\textcopyright} 2024 Copyright held by the owner/author(s).; 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024 ; Conference date: 29-10-2024 Through 01-11-2024",
year = "2024",
month = nov,
day = "22",
doi = "10.1145/3678717.3691262",
language = "English",
pages = "410--420",
editor = "Nascimento, {Mario A.} and Li Xiong and Andreas Zufle and Yao-Yi Chiang and Ahmed Eldawy and Peer Kroger",
booktitle = "32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems",

}

Download

TY - GEN

T1 - Adaptive Dispatching of Mobile Charging Stations using Multi-Agent Graph Convolutional Cooperative-Competitive Reinforcement Learning

AU - Wonsak, Shimon

AU - Henke, Nils

AU - Al-Rifai, Mohammad

AU - Nolting, Michael

AU - Nejdl, Wolfgang

N1 - Publisher Copyright: © 2024 Copyright held by the owner/author(s).

PY - 2024/11/22

Y1 - 2024/11/22

N2 - Battery electric vehicles (BEV) offer an opportunity to decrease transportation and mobility emissions significantly. The availability of charging station networks and infrastructure is crucial for the proliferation of BEVs. While the expansion of the charging networks is still slow, optimal utilization of the existing infrastructure and dispatching of mobile charging stations can serve as a bypass while more charging stations are built. In this work, we propose a novel multi-agent reinforcement learning - AdapMCS - approach for optimizing the adaptive dispatching of mobile charging stations to maximize the number of served charging requests by a charging station operator while improving the customer experience. By combining graph neural networks with reinforcement learning our approach is able to adapt to dynamic spatio-temporal changes in the demand distribution, for example, during big events such as concerts or fairs. Furthermore, we conduct a thorough evaluation using a publicly available real-world dataset and simulation of dynamic demand distribution changes. The results show that our adaptive dispatching approach is able to deal with the demand shifts and achieve significant gains for both customers, in terms of reducing waiting and charging times, and operators, in terms of increasing their profit.

AB - Battery electric vehicles (BEV) offer an opportunity to decrease transportation and mobility emissions significantly. The availability of charging station networks and infrastructure is crucial for the proliferation of BEVs. While the expansion of the charging networks is still slow, optimal utilization of the existing infrastructure and dispatching of mobile charging stations can serve as a bypass while more charging stations are built. In this work, we propose a novel multi-agent reinforcement learning - AdapMCS - approach for optimizing the adaptive dispatching of mobile charging stations to maximize the number of served charging requests by a charging station operator while improving the customer experience. By combining graph neural networks with reinforcement learning our approach is able to adapt to dynamic spatio-temporal changes in the demand distribution, for example, during big events such as concerts or fairs. Furthermore, we conduct a thorough evaluation using a publicly available real-world dataset and simulation of dynamic demand distribution changes. The results show that our adaptive dispatching approach is able to deal with the demand shifts and achieve significant gains for both customers, in terms of reducing waiting and charging times, and operators, in terms of increasing their profit.

KW - Graph Neural Networks

KW - Mobile Charging Stations

KW - Multi-Agent

KW - Reinforcement Learning

UR - http://www.scopus.com/inward/record.url?scp=85215088925&partnerID=8YFLogxK

U2 - 10.1145/3678717.3691262

DO - 10.1145/3678717.3691262

M3 - Conference contribution

AN - SCOPUS:85215088925

SP - 410

EP - 420

BT - 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems

A2 - Nascimento, Mario A.

A2 - Xiong, Li

A2 - Zufle, Andreas

A2 - Chiang, Yao-Yi

A2 - Eldawy, Ahmed

A2 - Kroger, Peer

T2 - 32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024

Y2 - 29 October 2024 through 1 November 2024

ER -

By the same author(s)