Applying Fuzzy Reliability Analysis of Damaged Road Network to Disaster Reduction Planning

Abstract

After the occurrence of a large-scale disaster, the assessment of road damage can only rely on the experts to judge according to their experience and judgment. This study uses linguistic variables to describe the link connectivity under damaged conditions. Link connectivity is estimated based on human subjective cognition of road damage. To assess the vulnerability of the road network, we propose two kinds of fuzzy reliability indicators, fuzzy reliability of travel time and fuzzy reliability of network connection. They are the decision basis for disaster reduction planning. Due to difficulties in obtaining actual data limited to large-scale disasters, this study was analyzed using the Sioux Falls road network. According to the probability of damage to the road affected by the earthquake, we randomly generate the road damage scenarios according to the road width and use four linguistic variables to describe the road connectivity, i.e."high", "medium", "low", and "very low". The simulation situation analysis results show that node 14, node 23 and node 24 located in the southwest corner of Sioux Falls road network are the most vulnerable. Once a large earthquake disaster occurs, the connection degree of external traffic is the lowest and should be improved first.