A COMPARATIVE ANALYSIS OF INCIDENT SERVICE TIME ON URBAN FREEWAYS* *This research was supported by the Korean Research Institute for Human Settlements (KRIHS)

The occurrence of an incident is one of the main causes of traffic congestion and delay on urban freeways. Currently, the development of an efficient traffic incident management process is an emerging issue in intelligent transport systems. The focus of this study is to develop an incident response model for use in an incident management program designed to reduce incident related traffic congestion and delays. Establishing a fuzzy control system based on a real operation process of incident response would help decrease the workload of operators who are constantly making essential decisions, and would be helpful in developing a more reliable operation procedure. Traffic incident data obtained from traffic detectors and information systems on freeways should be considered in order to determine the best form of operation. In this study, a fuzzy incident response model is formulated in terms of incident, type of vehicle, type of incident vehicle, location of incident vehicle, and incident service time. In order to analyze the reliability of the proposed model, the application of the model is made using the actual incident data collected on the freeway in the Los Angeles area. The application of the model shows that the fuzzy incident response system is very effective in describing the actual judgment of the incident operators in terms of incident service time.