Cascading failures and resilience evolution in urban road traffic networks with bounded rational route choice

Abstract

The growing complexity of urban road traffic networks (URTNs) has increased their vulnerability to cascading failures. As one of the fundamental sources of network dynamics, route choice behavior should be essential for the pattern of resilience against cascading failures, but the mechanism between behavioral uncertainty and resilience evolution remains understudied. This paper proposes a novel framework for analyzing URTN resilience under cascading failures, focusing on how behavioral uncertainty shapes failure process and resilience evolution. Specifically, we develop a link-based cascading failure model that integrates load redistribution processes with route choice behavior under uncertainty, incorporating dynamic capacity changes and event intensity effects. The model employs fuzzy cumulative prospect theory (FCPT) to capture travelers' decision-making characteristics, considering both risk preference and perception uncertainty. A multi-level resilience assessment indicator is introduced to evaluate network performance from both structural and operational perspectives. Using Shangyu's URTN as the case study, we find that: (1) the network exhibits moderately high resilience with failure propagation following distinct thresholds; (2) high-importance link failures trigger severe cascading effects, particularly during sustained area-wide disruptions and peak hours; (3) optimal network performance is achieved with a tolerance parameter of 0.2 and moderate risk preference combined with low-to-medium perception uncertainty; and (4) recovery strategies should prioritize directly attacked links to effectively mitigate cascade propagation. These findings provide valuable references for studying URTN resilience against unforeseen cascading failures.