Reliability Assessment of Road Network to Precipitation Based on Historical Recorded Disruptions

Road networks play an essential role in economic activities and are required to have high level of reliability. Heavy precipitation could lead to disruptions to road networks, causing service failure and associated economic and social impacts. In this study, we proposed a framework for assessing the reliability of road networks to precipitation based on historical recorded disruption data. We first developed a vulnerability disruption model, which describes the relationship between precipitation intensity and the probability of road disruption, as well as the amount of disruptions, based on recorded data. The vulnerability of roads shows a nonlinear increasing trend with increasing precipitation. Based on the vulnerability model and Monte Carlo approach, the travel time reliability of road network of Fujian province, China was assessed. Two precipitation events in Fujian road network were selected to validate the proposed framework. The travel time reliability of 11.12% city pairs in Fujian Province decreased to below 0.85 with an average precipitation of 73.32 mm. The proposed vulnerability model and the framework for assessing reliability can provide guidance to highway departments so that they can prepare for possible adverse impacts caused by precipitation on roads and plan for road network service.