pith. sign in

arxiv: 2605.16784 · v1 · pith:C7N3IKWKnew · submitted 2026-05-16 · 💻 cs.MA

Dynamic Deployment of Mobile Charging Trucks During Natural Disaster Evacuation: An Offline-to-Online Framework

Rui Ma , Zilin Bian , Kaan Ozbay This is my paper

Pith reviewed 2026-05-19 19:58 UTC · model grok-4.3

classification 💻 cs.MA
keywords mobile charging truckselectric vehicle evacuationrisk-aware deploymentoffline-to-online frameworkdynamic resource allocationdisaster responsenatural disaster evacuation
0
0 comments X

The pith

An adaptive offline-to-online framework deploys mobile charging trucks to cut electric vehicle risk exposure during evacuations by up to 71 percent in simulations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a framework to dynamically deploy mobile charging trucks that supplement fixed stations when concentrated demand during evacuations causes overloads and longer waits. It separates the problem into allocating trucks among stations and routing them, solving the allocation through a policy trained offline in a simulator and then refined online to match current conditions. A sympathetic reader would care because extended charging times can trap people in hazardous zones during hurricanes or similar events, and the method demonstrates clear risk reductions across demand shifts and infrastructure breakdowns. The results indicate that this hybrid training approach handles uncertainty more effectively than purely offline plans or simple real-time rules.

Core claim

The paper claims that its Adaptive Risk-aware MCT Deployment framework, which formulates truck allocation as a decentralized partially observable Markov decision process solved by a multi-agent policy pre-trained offline and adapted online while using a spatio-temporal predictor for rolling-horizon routing, consistently lowers average risk exposure relative to baselines without trucks, offline optimization alone, online heuristics, or rolling-horizon methods, reaching reductions of up to 71.1 percent under demand perturbations and 39.3 to 60.5 percent under fixed infrastructure or road failures in a Hillsborough County hurricane evacuation simulator.

What carries the argument

The Adaptive Risk-aware MCT Deployment (ARMD) offline-to-online framework that coordinates allocation of mobile charging trucks among fixed stations via learned decentralized policies and updates routes with predicted travel times.

If this is right

  • Evacuation planners obtain a method to reduce risk exposure by adding mobile assets that respond to changing demand and network conditions in real time.
  • The advantage of the adaptive approach grows as disruptions become more severe, indicating greater value precisely when fixed infrastructure is most stressed.
  • Multiple trucks can coordinate effectively even when each observes only partial information about the overall evacuation state.
  • Online policy refinement improves outcomes over static offline solutions or purely reactive heuristics during evolving events.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same offline-to-online structure could support deployment of other mobile resources, such as water tankers or medical units, in comparable emergency settings.
  • Feeding real-time vehicle location data into the travel-time predictor might further tighten routing decisions beyond the current simulator tests.
  • Applying the framework across multiple counties or different disaster types would reveal whether the risk reductions hold outside the original training region.

Load-bearing premise

The simulated hurricane evacuation environment built from Hillsborough County data sufficiently captures real-world EV charging demand patterns, travel times, and behavioral responses under uncertainty to allow reliable policy transfer from offline training to online operation.

What would settle it

A side-by-side comparison of measured risk exposure in a real-world evacuation against the simulator outputs, with and without the adaptive truck deployment, would confirm or refute whether the learned policies transfer effectively.

Figures

Figures reproduced from arXiv: 2605.16784 by Kaan Ozbay, Rui Ma, Zilin Bian.

Figure 1
Figure 1. Figure 1: Spatio-temporal demand–supply imbalance in charging network during evacuation, illus [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architecture of the ARMD framework. It divides MCT deployment into two coupled [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Study area in Hillsborough County, Florida: (a) study area; (b) locations of FCSs in study [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Station-level queue heatmaps over time under the base scenario. [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Temporal evolution of system risk across different scenarios: (a)-(c) evacuation demand [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Effects of road network fragmentation on EV travel distance and charging demand. [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Mean AFD and Gini coefficient of AFD under different scenarios. [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Sensitivity analysis of MCT fleet size under ARMD and Greedy. [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
read the original abstract

During large-scale evacuations, concentrated electric vehicle (EV) charging demand can overload fixed charging stations (FCSs), leading to prolonged waiting time and increased risk exposure. To address this challenge, this study proposes dynamically deploying mobile charging trucks (MCTs) to complement FCSs, and develops an Adaptive Risk-aware MCT Deployment (ARMD) framework for real-time operation. It divides the MCT deployment into two problems: risk-aware allocation of MCTs among FCSs and dynamic routing of MCTs to the assigned FCSs, and solves them under an offline-to-online paradigm. The resource allocation problem is formulated as a decentralized partially observable Markov decision process, and a multi-agent proximal policy optimization (MAPPO)-based policy is developed to coordinate multiple MCTs under decentralized observations. The policy is pre-trained offline in an evacuation simulator and adaptively refined online according to current evacuation context. For routing, a spatio-temporal travel time predictor is developed to support rolling-horizon route updates. The proposed framework is evaluated in a simulated hurricane evacuation environment built using real-world data from Hillsborough County, Florida. Experiments show that ARMD consistently outperforms offline optimization, online heuristic dispatch, and rolling-horizon optimization in reducing risk exposure. For demand perturbation scenarios, ARMD reduces average risk exposure by up to 71.1%, relative to the baseline without MCTs. In the case of fixed e-vehicle charging infrastructure or road link failures, ARMD achieves 39.3% to 60.5% reduction in average risk exposure, with its advantages becoming more pronounced as the severity of disruption increases. These results demonstrate the effectiveness and robustness of ARMD in enhancing mobile charging operations for realistic scenarios of uncertain evacuation conditions.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces the ARMD framework for dynamic deployment of mobile charging trucks (MCTs) during evacuations to reduce EV charging risk exposure. It formulates allocation as a decentralized POMDP solved via MAPPO with offline pre-training and online adaptation, paired with a spatio-temporal travel time predictor for routing. Evaluation in a Hillsborough County hurricane evacuation simulator shows ARMD outperforming baselines, achieving up to 71.1% risk reduction in demand perturbations and 39.3%-60.5% in infrastructure failure scenarios.

Significance. Should the simulator faithfully capture real evacuation dynamics, this work offers a promising offline-to-online multi-agent RL approach for resilient charging infrastructure in disasters. The combination of pre-trained policies with online refinement and predictive routing is a positive contribution to handling uncertainty in multi-agent systems for emergency response.

major comments (2)
  1. [Abstract and Evaluation] Abstract and Evaluation section: The central performance claims, including up to 71.1% and 39.3% to 60.5% reductions in average risk exposure, are obtained solely within the described simulator without reported validation against real-world evacuation data, statistical significance tests, or sensitivity analysis to parameters such as demand generation or behavioral responses. This is load-bearing for the quantitative improvements over the three baselines.
  2. [§3] §3 (Offline-to-Online Framework): The MAPPO policy is pre-trained in the simulator and refined online; the manuscript provides no analysis of policy transfer under simulator mismatch or sensitivity of the reported risk reductions to alternative behavioral or disruption parameters in the demand and travel-time modules.
minor comments (2)
  1. [Abstract] The abstract uses 'risk exposure' without a brief definition or reference to its computation formula.
  2. [Introduction] Consider adding citations to recent literature on EV charging demand modeling during evacuations.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below with clarifications and proposed revisions to improve the robustness and transparency of our evaluation.

read point-by-point responses

  1. Referee: [Abstract and Evaluation] Abstract and Evaluation section: The central performance claims, including up to 71.1% and 39.3% to 60.5% reductions in average risk exposure, are obtained solely within the described simulator without reported validation against real-world evacuation data, statistical significance tests, or sensitivity analysis to parameters such as demand generation or behavioral responses. This is load-bearing for the quantitative improvements over the three baselines.

    Authors: We acknowledge that all quantitative claims are obtained from the Hillsborough County simulator parameterized with real-world road networks, charging infrastructure, and historical evacuation data. Direct validation against observed real-world evacuation outcomes is not reported, as such granular post-event data for EV charging behavior during hurricanes is not publicly available for this study. However, we agree that statistical significance testing and sensitivity analysis are feasible additions. In the revised manuscript, we will report paired t-tests or Wilcoxon tests on the risk reduction metrics across multiple simulation runs and add sensitivity analyses varying demand generation rates and behavioral response parameters (e.g., evacuation compliance rates). We will also expand the limitations section to discuss simulator fidelity more explicitly. revision: partial

  2. Referee: [§3] §3 (Offline-to-Online Framework): The MAPPO policy is pre-trained in the simulator and refined online; the manuscript provides no analysis of policy transfer under simulator mismatch or sensitivity of the reported risk reductions to alternative behavioral or disruption parameters in the demand and travel-time modules.

    Authors: We appreciate this point on robustness. The current experiments already include demand perturbation and infrastructure failure scenarios, but we agree that explicit analysis of policy transfer under simulator mismatch and broader sensitivity to alternative parameters would strengthen the offline-to-online claims. In the revision, we will add experiments evaluating the pre-trained MAPPO policy under mismatched conditions (e.g., perturbed travel-time distributions and alternative demand models) and report sensitivity of risk reductions to variations in the spatio-temporal predictor parameters and behavioral assumptions in the demand module. revision: yes

standing simulated objections not resolved
  • Direct validation of simulator outputs against real-world evacuation data, which would require detailed empirical observations of EV charging and routing behavior during actual hurricane events not available for this work.

Circularity Check

0 steps flagged

No significant circularity in derivation chain or performance claims

full rationale

The paper describes an offline-to-online ARMD framework using MAPPO for MCT allocation and a travel-time predictor for routing, with all quantitative results (71.1% and 39.3–60.5% risk reductions) obtained as direct empirical outputs from running the trained policy and baselines inside a single simulator built on Hillsborough County data. No mathematical derivation, equation, or policy step is shown to reduce by construction to a fitted parameter, self-defined quantity, or load-bearing self-citation. The simulator functions as an external benchmark environment for evaluation rather than embedding the target result in its definition. Standard RL training and rolling-horizon updates do not trigger any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review prevents enumeration of specific free parameters or simulator assumptions; the MAPPO policy and travel-time predictor almost certainly contain hyperparameters and modeling choices whose values are not reported here.

pith-pipeline@v0.9.0 · 5849 in / 1108 out tokens · 41906 ms · 2026-05-19T19:58:31.900544+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

62 extracted references · 62 canonical work pages · 1 internal anchor

  1. [1]

    Evacuation route planning for alternative fuel vehicles.Transportation research part C: emerging technologies, 143:103837, 2022

    Denissa Sari Darmawi Purba, Eleftheria Kontou, and Chrysafis V ogiatzis. Evacuation route planning for alternative fuel vehicles.Transportation research part C: emerging technologies, 143:103837, 2022

    work page 2022

  2. [2]

    Assessing the impact of electric vehicles on traffic emissions: An agent-based modeling approach considering traveler behavior changes.Procedia Computer Science, 257: 329–335, 2025

    Ding Wang, Mohammad Tayarani, Jingqin Gao, Zilin Bian, Kaan Ozbay, H Oliver Gao, and Joseph YJ Chow. Assessing the impact of electric vehicles on traffic emissions: An agent-based modeling approach considering traveler behavior changes.Procedia Computer Science, 257: 329–335, 2025

    work page 2025

  3. [3]

    Zerun Liu, Tu Lan, Zilin Bian, Jingqin Gao, and Kaan Ozbay. A comprehensive framework for the assessment of the effects of increased electric truck weights on road infrastructure: A new york city case study.Transport Policy, page 103808, 2025

    work page 2025

  4. [4]

    Department of Energy

    U.S. Department of Energy. All-Electric Vehicles Had a Median Driving Range about 60% That of Gasoline Powered Vehicles. https://www.energy.gov/eere/vehicles/articles/ fotw-1221-january-17-2022-model-year-2021-all-electric-vehicles-had-median ,

    work page 2022

  5. [6]

    Gas Stations in the United States of America: Ev- erything You Need to Know

    xMap.ai. Gas Stations in the United States of America: Ev- erything You Need to Know. https://www.xmap.ai/blog/ gas-stations-in-united-states-of-america-everything-you-need-to-know ,

    work page

  6. [8]

    Department of Energy

    U.S. Department of Energy. Electric Vehicle Charging Station Locations. https://afdc. energy.gov/fuels/electricity_locations.html#/find/nearest?fuel=ELEC,

    work page

  7. [9]

    Accessed: 2026-03-02

    work page 2026

  8. [10]

    Department of Transportation

    U.S. Department of Transportation. Charger types and speeds. https://www. transportation.gov/rural/ev/toolkit/ev-basics/charging-speeds, 2025. Ac- cessed: 2026-03-02

    work page 2025

  9. [11]

    Eiman Elghanam, Akmal Abdelfatah, Mohamed S Hassan, and Ahmed H Osman. Optimiza- tion techniques in electric vehicle charging scheduling, routing and spatio-temporal demand coordination: A systematic review.IEEE Open Journal of Vehicular Technology, 5:1294–1313, 2024

    work page 2024

  10. [12]

    Reinforcement learning for electric vehicle charging scheduling: A systematic review.Transportation Research Part E: Logistics and Transportation Review, 190:103698, 2024

    Zhonghao Zhao, Carman KM Lee, Xiaoyuan Yan, and Haonan Wang. Reinforcement learning for electric vehicle charging scheduling: A systematic review.Transportation Research Part E: Logistics and Transportation Review, 190:103698, 2024

    work page 2024

  11. [13]

    Charging station location problem: A comprehensive review on models and solution approaches.Transportation Research Part C: Emerging Technologies, 132:103376, 2021

    Mouna Kchaou-Boujelben. Charging station location problem: A comprehensive review on models and solution approaches.Transportation Research Part C: Emerging Technologies, 132:103376, 2021. ISSN 0968-090X. doi: https://doi.org/10.1016/j.trc.2021.103376. URL https://www.sciencedirect.com/science/article/pii/S0968090X21003776

    work page doi:10.1016/j.trc.2021.103376 2021

  12. [15]

    Optimal planning of flood-resilient electric vehicle charging stations.Computer-Aided Civil and Infrastructure Engineering, 38(4): 489–507, 2023

    Qian Zhang, Hao Yu, Guohui Zhang, and Tianwei Ma. Optimal planning of flood-resilient electric vehicle charging stations.Computer-Aided Civil and Infrastructure Engineering, 38(4): 489–507, 2023

    work page 2023

  13. [16]

    Optimal energy management framework for truck-mounted mobile charging stations considering power distribution system operating conditions.Sensors, 21(8):2798, 2021

    Soi Jeon and Dae-Hyun Choi. Optimal energy management framework for truck-mounted mobile charging stations considering power distribution system operating conditions.Sensors, 21(8):2798, 2021. 21

    work page 2021

  14. [17]

    Florida Deploys Emergency Mobile Electric Vehicle Charging Stations Along Hurricane Evacuation Routes

    Joint Office of Energy and Transportation. Florida Deploys Emergency Mobile Electric Vehicle Charging Stations Along Hurricane Evacuation Routes. https://driveelectric.gov/ news/florida-mobile-charging, 2024. Accessed: 2026-03-02

    work page 2024

  15. [18]

    Can we evacuate from hurricanes with electric vehicles?Transportation research part D: transport and environment, 86:102458, 2020

    Kairui Feng, Ning Lin, Siyuan Xian, and Mikhail V Chester. Can we evacuate from hurricanes with electric vehicles?Transportation research part D: transport and environment, 86:102458, 2020

    work page 2020

  16. [19]

    Hidden benefits of electric vehicles for addressing climate change.Scientific reports, 5(1):9213, 2015

    Canbing Li, Yijia Cao, Mi Zhang, Jianhui Wang, Jianguo Liu, Haiqing Shi, and Yinghui Geng. Hidden benefits of electric vehicles for addressing climate change.Scientific reports, 5(1):9213, 2015

    work page 2015

  17. [20]

    Temperature effects on electric and hybrid vehicle efficiency: Dynamometer-based efficiency comparison of hybrid and battery electric vehicles

    American Automobile Association. Temperature effects on electric and hybrid vehicle efficiency: Dynamometer-based efficiency comparison of hybrid and battery electric vehicles. Technical report, American Automobile Association, May 2026. URL https://newsroom.aaa.com. Accessed: 2026-05-13

    work page 2026

  18. [21]

    Electric vehicles and natural disaster policy implications.Energy Policy, 112:437–448, 2018

    Shawn A Adderly, Daria Manukian, Timothy D Sullivan, and Mun Son. Electric vehicles and natural disaster policy implications.Energy Policy, 112:437–448, 2018

    work page 2018

  19. [22]

    Optimal mass evacuation planning for electric vehicles before natural disasters.Transportation research part D: transport and environment, 107:103292, 2022

    Qianwen Li, Saeid Soleimaniamiri, and Xiaopeng Li. Optimal mass evacuation planning for electric vehicles before natural disasters.Transportation research part D: transport and environment, 107:103292, 2022

    work page 2022

  20. [23]

    Emergency management plan for electric vehicles during floods using daily routing patterns

    Alaa Torkey, Mohamed H Zaki, and Ashraf A El Damatty. Emergency management plan for electric vehicles during floods using daily routing patterns. In2024 IEEE International Conference on Smart Mobility (SM), pages 183–188. IEEE, 2024

    work page 2024

  21. [24]

    A multi-trip electric bus routing model considering equity during short-notice evacuations.Transportation Research Part D: Transport and Environment, 110:103397, 2022

    Jingwen Zhang and Xiaoning Zhang. A multi-trip electric bus routing model considering equity during short-notice evacuations.Transportation Research Part D: Transport and Environment, 110:103397, 2022

    work page 2022

  22. [25]

    Resilience oriented planning of distribution system with electric vehicle charging station.IEEE Transactions on Industry Applications, 60(2):2214–2224, 2023

    Yuwei Chong, Wen-Shan Tan, Susilawati Susilawati, Ming Fai Chow, and Yuan-Kang Wu. Resilience oriented planning of distribution system with electric vehicle charging station.IEEE Transactions on Industry Applications, 60(2):2214–2224, 2023

    work page 2023

  23. [26]

    Feng Chen, Yang Peng, Bing Han, Shaofeng Lu, Fei Xue, and Gan Li. Robust planning of electric vehicle charging stations considering demand uncertainty and facility failures.IEEE Transactions on Transportation Electrification, 10(3):7551–7564, 2023

    work page 2023

  24. [27]

    Refueling station location model to support evacuation of alternative fuel vehicles.Transportation Research Record, 2678 (1):521–538, 2024

    Denissa Sari Darmawi Purba, Simon Balisi, and Eleftheria Kontou. Refueling station location model to support evacuation of alternative fuel vehicles.Transportation Research Record, 2678 (1):521–538, 2024

    work page 2024

  25. [28]

    Measuring the effectiveness of incorporating mobile charging services into urban electric vehicle charging network: An agent-based approach.Renewable Energy, 234:121246, 2024

    Bingkun Chen, Zhuo Chen, Xiaoyue Cathy Liu, Nan Zheng, and Qijie Xiao. Measuring the effectiveness of incorporating mobile charging services into urban electric vehicle charging network: An agent-based approach.Renewable Energy, 234:121246, 2024

    work page 2024

  26. [29]

    Mobile charging stations for electric vehicles—a review.Renewable and Sustainable Energy Reviews, 152:111654, 2021

    Shahab Afshar, Pablo Macedo, Farog Mohamed, and Vahid Disfani. Mobile charging stations for electric vehicles—a review.Renewable and Sustainable Energy Reviews, 152:111654, 2021

    work page 2021

  27. [30]

    Mobile charging: A novel charging system for electric vehicles in urban areas.Applied Energy, 278: 115648, 2020

    Yaoli Zhang, Xingyu Liu, Wenshen Wei, Tianji Peng, Gang Hong, and Chao Meng. Mobile charging: A novel charging system for electric vehicles in urban areas.Applied Energy, 278: 115648, 2020

    work page 2020

  28. [31]

    Hybrid of fixed and mobile charging systems for electric vehicles: System design and analysis.Transportation Research Part C: Emerging Technologies, 126:103068, 2021

    Chengzhang Wang, Xi Lin, Fang He, Max Zuo-jun Shen, and Meng Li. Hybrid of fixed and mobile charging systems for electric vehicles: System design and analysis.Transportation Research Part C: Emerging Technologies, 126:103068, 2021

    work page 2021

  29. [32]

    Online-to-offline mobile charging system for electric vehicles: Strategic planning and online operation.Transportation Research Part D: Transport and Environment, 87:102522, 2020

    Peng Tang, Fang He, Xi Lin, and Meng Li. Online-to-offline mobile charging system for electric vehicles: Strategic planning and online operation.Transportation Research Part D: Transport and Environment, 87:102522, 2020. 22

    work page 2020

  30. [33]

    The multi-mode mobile charging service based on electric vehicle spatiotemporal distribution.Energy, 198:117302, 2020

    Shaohua Cui, Baozhen Yao, Gang Chen, Chao Zhu, and Bin Yu. The multi-mode mobile charging service based on electric vehicle spatiotemporal distribution.Energy, 198:117302, 2020

    work page 2020

  31. [34]

    Joint routing and scheduling of mobile charging infrastructure for v2v energy transfer.IEEE Transactions on Intelligent Vehicles, 6(4):736–746, 2021

    Mohammad Ekramul Kabir, Ibrahim Sorkhoh, Bassam Moussa, and Chadi Assi. Joint routing and scheduling of mobile charging infrastructure for v2v energy transfer.IEEE Transactions on Intelligent Vehicles, 6(4):736–746, 2021

    work page 2021

  32. [35]

    Mobile charging stations for ev charging management in urban areas: A case study in chattanooga.Applied Energy, 325:119901, 2022

    Shahab Afshar, Zachary K Pecenak, Masoud Barati, and Vahid Disfani. Mobile charging stations for ev charging management in urban areas: A case study in chattanooga.Applied Energy, 325:119901, 2022

    work page 2022

  33. [36]

    Dynamic routing and scheduling of mobile charging stations for electric vehicles using deep reinforcement learning

    Ubaid Qureshi, Arnob Ghosh, and BK Panigrahi. Dynamic routing and scheduling of mobile charging stations for electric vehicles using deep reinforcement learning. In2024 IEEE Power & Energy Society General Meeting (PESGM), pages 1–5. IEEE, 2024

    work page 2024

  34. [37]

    Electric vehicle charging optimization with coordinated mobile and fixed chargers.Transportation Research Part E: Logistics and Transportation Review, 204:104434, 2025

    Xiaofeng Li, Xinlian Yu, Ziyuan Pu, and Jingxu Chen. Electric vehicle charging optimization with coordinated mobile and fixed chargers.Transportation Research Part E: Logistics and Transportation Review, 204:104434, 2025

    work page 2025

  35. [38]

    Coordinated scheduling of ev charging service and energy arbitrage for truck mobile charging stations.IEEE Transactions on Smart Grid, 2025

    Kecheng He, Hongjie Jia, Yunfei Mu, Xiaodan Yu, Yue Zhou, Wei Gan, and Jianzhong Wu. Coordinated scheduling of ev charging service and energy arbitrage for truck mobile charging stations.IEEE Transactions on Smart Grid, 2025

    work page 2025

  36. [39]

    Mobile charging station placements in internet of electric vehicles: A federated learning approach.IEEE Transactions on Intelligent Transportation Systems, 23(12):24561–24577, 2022

    Linfeng Liu, Zhiyuan Xi, Kun Zhu, Ran Wang, and Ekram Hossain. Mobile charging station placements in internet of electric vehicles: A federated learning approach.IEEE Transactions on Intelligent Transportation Systems, 23(12):24561–24577, 2022

    work page 2022

  37. [40]

    Linfeng Liu, Su Liu, Jiagao Wu, and Jia Xu. A placement strategy for idle mobile charging stations in ioev: From the view of charging demand force.IEEE Transactions on Intelligent Transportation Systems, 25(5):3870–3884, 2023

    work page 2023

  38. [41]

    NIO Power Mobile

    NIO. NIO Power Mobile. https://www.nio.com/videos/nio-power-mobile, 2020. Accessed: 2026-01-03

    work page 2020

  39. [42]

    Tesla Rolls Out Megapack-Chargers to Ease Holiday EV Charging Surge

    Michael Phoon. Tesla Rolls Out Megapack-Chargers to Ease Holiday EV Charging Surge. https://ev.com/news/ tesla-rolls-out-megapackchargers-to-ease-holiday-ev-charging-surge ,

    work page

  40. [43]

    Accessed: 2026-01-03

    work page 2026

  41. [44]

    An optimization model for the temporary locations of mobile charging stations.Mathematics, 8(3):453, 2020

    Maria-Simona R˘aboac˘a, Irina B˘ancescu, Vasile Preda, and Nicu Bizon. An optimization model for the temporary locations of mobile charging stations.Mathematics, 8(3):453, 2020

    work page 2020

  42. [45]

    Dispatch management of portable charging stations in electric vehicle networks.ETransportation, 8: 100112, 2021

    Valeh Moghaddam, Iftekhar Ahmad, Daryoush Habibi, and Mohammad AS Masoum. Dispatch management of portable charging stations in electric vehicle networks.ETransportation, 8: 100112, 2021

    work page 2021

  43. [46]

    Ubaid Qureshi, Arnob Ghosh, and Bijaya Ketan Panigrahi. Scheduling and routing of mobile charging stations with stochastic travel times to service heterogeneous spatiotemporal electric vehicle charging requests with time windows.IEEE Transactions on Industry Applications, 58 (5):6546–6556, 2022

    work page 2022

  44. [47]

    Ali Ala, Muhammet Deveci, Erfan Amani Bani, and Amir Hossein Sadeghi. Dynamic capaci- tated facility location problem in mobile renewable energy charging stations under sustainability consideration.Sustainable Computing: Informatics and Systems, 41:100954, 2024

    work page 2024

  45. [48]

    Probabilistic programming models for traffic incident management operations planning.Annals of Operations Research, 203(1):389–406, 2013

    Kaan Ozbay, Cem Iyigun, Melike Baykal-Gursoy, and Weihua Xiao. Probabilistic programming models for traffic incident management operations planning.Annals of Operations Research, 203(1):389–406, 2013

    work page 2013

  46. [49]

    The surprising effectiveness of ppo in cooperative multi-agent games.Advances in neural information processing systems, 35:24611–24624, 2022

    Chao Yu, Akash Velu, Eugene Vinitsky, Jiaxuan Gao, Yu Wang, Alexandre Bayen, and Yi Wu. The surprising effectiveness of ppo in cooperative multi-agent games.Advances in neural information processing systems, 35:24611–24624, 2022. 23

    work page 2022

  47. [50]

    Complementary attention for multi-agent reinforcement learning

    Jianzhun Shao, Hongchang Zhang, Yun Qu, Chang Liu, Shuncheng He, Yuhang Jiang, and Xiangyang Ji. Complementary attention for multi-agent reinforcement learning. InInternational conference on machine learning, pages 30776–30793. PMLR, 2023

    work page 2023

  48. [51]

    Active advantage-aligned online reinforcement learning with offline data.arXiv preprint arXiv:2502.07937,

    Xuefeng Liu, Hung TC Le, Siyu Chen, Rick Stevens, Zhuoran Yang, Matthew R Walter, and Yuxin Chen. Active advantage-aligned online reinforcement learning with offline data.arXiv preprint arXiv:2502.07937, 2025

    work page arXiv 2025

  49. [52]

    Semi-Supervised Classification with Graph Convolutional Networks

    Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks.arXiv preprint arXiv:1609.02907, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  50. [53]

    Tao Li, Zilin Bian, Haozhe Lei, Fan Zuo, Ya-Ting Yang, Quanyan Zhu, Zhenning Li, Zhibin Chen, and Kaan Ozbay. Digital twin-based driver risk-aware predictive mobility analytics for real-time situational awareness through cooperative sensing.IEEE Transactions on Intelligent Transportation Systems, 2025

    work page 2025

  51. [54]

    Multi-level traffic-responsive tilt camera surveillance through predictive correlated online learning.Transportation Research Part C: Emerging Technologies, 167:104804, 2024

    Tao Li, Zilin Bian, Haozhe Lei, Fan Zuo, Ya-Ting Yang, Quanyan Zhu, Zhenning Li, and Kaan Ozbay. Multi-level traffic-responsive tilt camera surveillance through predictive correlated online learning.Transportation Research Part C: Emerging Technologies, 167:104804, 2024

    work page 2024

  52. [55]

    Evacuation zone map, n.d

    Hillsborough County Open Data Portal. Evacuation zone map, n.d. URL https:// gis2017-01-10t133755357z-hillsborough.opendata.arcgis.com/ . ArcGIS Open Data Portal, accessed 2026-03-31

    work page 2026

  53. [56]

    Department of Energy

    U.S. Department of Energy. Alternative fueling station locator, n.d.. URL https://afdc. energy.gov/stations. Alternative Fuels Data Center, accessed 2026-03-31

    work page 2026

  54. [57]

    Microscopic traffic simulation using sumo

    Pablo Alvarez Lopez, Michael Behrisch, Laura Bieker-Walz, Jakob Erdmann, Yun-Pang Flöt- teröd, Robert Hilbrich, Leonhard Lücken, Johannes Rummel, Peter Wagner, and Evamarie Wießner. Microscopic traffic simulation using sumo. In2018 21st international conference on intelligent transportation systems (ITSC), pages 2575–2582. Ieee, 2018

    work page 2018

  55. [58]

    Di Sha, Jingqin Gao, Di Yang, Fan Zuo, and Kaan Ozbay. Calibrating stochastic traffic simulation models for safety and operational measures based on vehicle conflict distributions obtained from aerial and traffic camera videos.Accident Analysis & Prevention, 179:106878, 2023

    work page 2023

  56. [59]

    Household evacuation decision making in response to hurricane ike.Natural Hazards Review, 13(4):283–296, 2012

    Shih-Kai Huang, Michael K Lindell, Carla S Prater, Hao-Che Wu, and Laura K Siebeneck. Household evacuation decision making in response to hurricane ike.Natural Hazards Review, 13(4):283–296, 2012

    work page 2012

  57. [60]

    Census Bureau

    U.S. Census Bureau. 2020 census demographic and housing character- istics, 2020. URL https://www.census.gov/data/tables/2020/dec/ 2020-census-demographic-and-housing-characteristics.html . accessed 2026-03- 31

    work page 2020

  58. [61]

    Compound risks of hurricane evacuation amid the covid-19 pandemic in the united states.GeoHealth, 4 (12):e2020GH000319, 2020

    Sen Pei, Kristina A Dahl, Teresa K Yamana, Rachel Licker, and Jeffrey Shaman. Compound risks of hurricane evacuation amid the covid-19 pandemic in the united states.GeoHealth, 4 (12):e2020GH000319, 2020

    work page 2020

  59. [62]

    Evacuation modelling in the united states: Does the demand model choice matter?Transport Reviews, 28(6):757–779, 2008

    Mustafa Anil Yazici and Kaan Ozbay. Evacuation modelling in the united states: Does the demand model choice matter?Transport Reviews, 28(6):757–779, 2008

    work page 2008

  60. [63]

    Department of Energy

    U.S. Department of Energy. Vehicle registration counts by state, n.d.. URL https://afdc. energy.gov/vehicle-registration. Alternative Fuels Data Center, accessed 2026-03-31

    work page 2026

  61. [64]

    Rethinking digital twin: Introducing digital risk twin for disaster risk management.npj Natural Hazards, 2(1):79, 2025

    Saman Ghaffarian. Rethinking digital twin: Introducing digital risk twin for disaster risk management.npj Natural Hazards, 2(1):79, 2025

    work page 2025

  62. [65]

    Digital post-disaster risk management twinning: A review and improved conceptual framework.International Journal of Disaster Risk Reduction, 110: 104629, 2024

    Umut Lagap and Saman Ghaffarian. Digital post-disaster risk management twinning: A review and improved conceptual framework.International Journal of Disaster Risk Reduction, 110: 104629, 2024. 24 A MIP Model This appendix presents the MIP model used by the OF-MIP and RH-MIP benchmarks. Let P= {0,1, . . . , M−1} denote the set of decision epochs covered in...

    work page 2024