Abstract

The transportation network resiliency plays a significant role in reducing the devastating impacts of a disaster. This paper develops a mathematical model to improve the resilience of road–bridge transportation networks in the recovery phase. The model consists of two objective functions. The first objective function minimises total recovery time, and the second one prioritises damaged bridges for restoration to increase network performance level in an early stage by maximising the skewness of the recovery trajectory. This paper considers disrupted bridges as repair projects and schedules a network of recovery tasks with precedence relationships to restore each bridge while considering resource constraints. Thus, it is similar to the resource–constraint multi-project scheduling problem. A genetic algorithm is applied to solve the model for a large-scale hypothetical road–bridge transport network, and the highway network of Shelby County, TN, the USA, following a seismic hazard. Different recovery task networks are considered to restore disrupted bridges. Three communities with rich, average and poor resources are compared to evaluate the impact of different community investment structures on the recovery process and network recovery time. The results show that the proposed restoration plan leads to achieving the best recovery trajectory and shortest total recovery time.

摘要

交通网络的弹性在减少灾难的破坏性影响方面发挥着重要作用。本文开发了一个数学模型，以提高路桥交通网络在恢复阶段的弹性。该模型由两个目标函数组成。第一个目标函数最小化总恢复时间，第二个目标函数优先考虑损坏的桥梁进行恢复，以通过最大化恢复轨迹的偏度来提高早期网络性能水平。本文将中断的桥梁视为修复项目，并安排具有优先关系的恢复任务网络，以在考虑资源限制的同时恢复每座桥梁。因此，它类似于资源约束的多项目调度问题。应用遗传算法求解大型假想路桥交通网络模型，以及美国田纳西州谢尔比县公路网络在地震灾害后的模型。考虑不同的恢复任务网络来恢复中断的网桥。通过比较资源丰富、平均和贫乏的三个社区，评估不同社区投资结构对恢复过程和网​​络恢复时间的影响。结果表明，所提出的恢复计划导致实现最佳恢复轨迹和最短的总恢复时间。比较平均资源和较差资源，评估不同社区投资结构对恢复过程和网​​络恢复时间的影响。结果表明，所提出的恢复计划导致实现最佳恢复轨迹和最短的总恢复时间。比较平均资源和较差资源，评估不同社区投资结构对恢复过程和网​​络恢复时间的影响。结果表明，所提出的恢复计划导致实现最佳恢复轨迹和最短的总恢复时间。