Digital Twin-Driven Dynamic Optimization for Urban Emergency Distribution
Liu Huwei
Liang Kaibo
Yang Jianglong
Zhao Junhui
Abstract：Urban emergency logistics systems face multiple challenges under sudden disaster scenarios,including dynamic demand evolution,real-time fluctuations in road network capacity,and limited inventory resources.Traditional static optimization methods struggle to effectively address environmental uncertainties.To address these issues,this paper proposes a digital twin-driven dynamic optimization approach for emergency distribution and constructs an integrated decision-making framework combining digital twin,optimization modeling,and reinforcement learning.The proposed method comprises three core components.First,a digital twin architecture for urban emergency logistics is established for real-time perception and virtual mapping of physical system states through multi-source data fusion.Second,a multi-period emergency distribution scheduling model considering dynamic inventory constraints is developed.Inventory balance equations are introduced to establish cross-period coupling relationships between distribution decisions and warehousing states,with the objective of minimizing the weighted response time.Third,an adaptive decision-making algorithm based on Deep Q-Network(DQN)is designed.The computational complexity is effectively reduced through state feature selection and action discretization strategies,enabling real-time plan adjustment in dynamic environments.An empirical study is conducted based on the"7·20"extreme rainstorm disaster in Zhengzhou,China.Results demonstrate that,compared with sequential decision-making methods,the proposed dynamic optimization model reduces weighted response time by 21.9％.The digital twin-based real-time decision algorithm maintains a solution feasibility rate of 92.3％in dynamic environments.Moreover,the DQN-based algorithm achieves a 23.2-fold improvement in computational efficiency compared to exact solution methods,with the acceleration ratio reaching up to 45.1 times as the problem scale increases.The research findings provide theoretical support and methodological reference for intelligent management of urban emergency logistics.
Keywords：emergency logisticsdigital twindistribution optimizationreinforcement learningdynamic decision-ma-king
Publication Date：2026-02-28
Online Publishing Date：2026-03-19(First online date of this platform, not the publication date of the document)
Pages：14( 61-74 )