Scholarlink

<p>The escalating frequency of extreme weather events poses significant challenges to the operational resilience of urban transportation networks. While existing recovery strategies often prioritize either structural integrity or operational performance, few offer adaptive mechanisms that dynamically balance these objectives throughout the recovery process. To address this gap, this study proposes a dual-layer “transportation network–flow network” architecture, establishes a unified framework for resilience assessment and decision-making, and employs a dynamic weighting mechanism based on a Sigmoid function to ensure that the focus of recovery shifts gradually from structural repair to service optimization as the recovery process progresses. Furthermore, we formulate the node recovery sequencing problem as a Markov Decision Process (MDP) and implement a Q-learning algorithm to identify near-optimal restoration policies. The proposed framework is validated through a real-world case study of the Zhengzhou “6.17” extreme rainstorm event. Results demonstrate that our method outperforms conventional static strategies, achieving a 24.50% improvement in network resilience under storm conditions. This work provides a theoretically grounded and computationally efficient tool for enhancing post-disaster recovery outcomes, with implications for urban transportation management and resilient infrastructure planning.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An emergency recovery strategy for road network resilience based on dynamic structure-function adaptive trade-offs

  • 期刊论文

  • 发表日期：

    2026年5月16日

查看全文
  • Qin Hongbo,
  • Cao Taiyu,
  • Li Penglong,
  • Zhu Yongming

摘要

The escalating frequency of extreme weather events poses significant challenges to the operational resilience of urban transportation networks. While existing recovery strategies often prioritize either structural integrity or operational performance, few offer adaptive mechanisms that dynamically balance these objectives throughout the recovery process. To address this gap, this study proposes a dual-layer “transportation network–flow network” architecture, establishes a unified framework for resilience assessment and decision-making, and employs a dynamic weighting mechanism based on a Sigmoid function to ensure that the focus of recovery shifts gradually from structural repair to service optimization as the recovery process progresses. Furthermore, we formulate the node recovery sequencing problem as a Markov Decision Process (MDP) and implement a Q-learning algorithm to identify near-optimal restoration policies. The proposed framework is validated through a real-world case study of the Zhengzhou “6.17” extreme rainstorm event. Results demonstrate that our method outperforms conventional static strategies, achieving a 24.50% improvement in network resilience under storm conditions. This work provides a theoretically grounded and computationally efficient tool for enhancing post-disaster recovery outcomes, with implications for urban transportation management and resilient infrastructure planning.