Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model

Emergency logistics is essential for rapid and efficient disaster response, ensuring the timely availability and deployment of resources to affected areas. In the process of rescue work, the dynamic changes in rescue point information greatly increase the difficulty of rescue. This paper establishes...

Full description

Saved in:
Bibliographic Details
Main Authors: Jin Zhang, Hao Xu, Ding Liu, Qi Yu
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Systems
Subjects:
Online Access:https://www.mdpi.com/2079-8954/13/2/127
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849720066583560192
author Jin Zhang
Hao Xu
Ding Liu
Qi Yu
author_facet Jin Zhang
Hao Xu
Ding Liu
Qi Yu
author_sort Jin Zhang
collection DOAJ
description Emergency logistics is essential for rapid and efficient disaster response, ensuring the timely availability and deployment of resources to affected areas. In the process of rescue work, the dynamic changes in rescue point information greatly increase the difficulty of rescue. This paper establishes a combined neural network model considering soft time-window penalty and applies deep reinforcement learning (DRL) to address the dynamic routing problem in emergency logistics. This method utilizes the actor–critic framework, combined with attention mechanisms, pointer networks, and long short-term memory neural networks, to determine effective disaster relief path, and it compares the obtained scheduling scheme with the results obtained from the DRL algorithm based on the single-network model and ant colony optimization (ACO) algorithm. Simulation experiments show that the proposed method reduces the solution accuracy by nearly 10% compared to the ACO algorithm, but it saves nearly 80% in solution time. Additionally, it slightly increases solution times but improves accuracy by nearly 20% over traditional DRL approaches, demonstrating a promising balance between performance efficiency and computational resource utilization in emergency logistics.
format Article
id doaj-art-4f8b33729ae846b3bd87311d69ecdb5b
institution DOAJ
issn 2079-8954
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Systems
spelling doaj-art-4f8b33729ae846b3bd87311d69ecdb5b2025-08-20T03:12:01ZengMDPI AGSystems2079-89542025-02-0113212710.3390/systems13020127Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning ModelJin Zhang0Hao Xu1Ding Liu2Qi Yu3College of Transport & Communications, Shanghai Maritime University, Shanghai 201306, ChinaCollege of Transport & Communications, Shanghai Maritime University, Shanghai 201306, ChinaCollege of Transport & Communications, Shanghai Maritime University, Shanghai 201306, ChinaCollege of Transportation Engineering, Tongji University, Shanghai 200092, ChinaEmergency logistics is essential for rapid and efficient disaster response, ensuring the timely availability and deployment of resources to affected areas. In the process of rescue work, the dynamic changes in rescue point information greatly increase the difficulty of rescue. This paper establishes a combined neural network model considering soft time-window penalty and applies deep reinforcement learning (DRL) to address the dynamic routing problem in emergency logistics. This method utilizes the actor–critic framework, combined with attention mechanisms, pointer networks, and long short-term memory neural networks, to determine effective disaster relief path, and it compares the obtained scheduling scheme with the results obtained from the DRL algorithm based on the single-network model and ant colony optimization (ACO) algorithm. Simulation experiments show that the proposed method reduces the solution accuracy by nearly 10% compared to the ACO algorithm, but it saves nearly 80% in solution time. Additionally, it slightly increases solution times but improves accuracy by nearly 20% over traditional DRL approaches, demonstrating a promising balance between performance efficiency and computational resource utilization in emergency logistics.https://www.mdpi.com/2079-8954/13/2/127emergency logisticsdynamic route optimizationdeep reinforcement learningcomposite networkvehicle routing problem
spellingShingle Jin Zhang
Hao Xu
Ding Liu
Qi Yu
Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
Systems
emergency logistics
dynamic route optimization
deep reinforcement learning
composite network
vehicle routing problem
title Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
title_full Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
title_fullStr Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
title_full_unstemmed Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
title_short Advancing Dynamic Emergency Route Optimization with a Composite Network Deep Reinforcement Learning Model
title_sort advancing dynamic emergency route optimization with a composite network deep reinforcement learning model
topic emergency logistics
dynamic route optimization
deep reinforcement learning
composite network
vehicle routing problem
url https://www.mdpi.com/2079-8954/13/2/127
work_keys_str_mv AT jinzhang advancingdynamicemergencyrouteoptimizationwithacompositenetworkdeepreinforcementlearningmodel
AT haoxu advancingdynamicemergencyrouteoptimizationwithacompositenetworkdeepreinforcementlearningmodel
AT dingliu advancingdynamicemergencyrouteoptimizationwithacompositenetworkdeepreinforcementlearningmodel
AT qiyu advancingdynamicemergencyrouteoptimizationwithacompositenetworkdeepreinforcementlearningmodel