This paper explores the feasibility of applying Artificial Intelligence (AI) techniques for urban network incident detection and control. It highlights progress made in the area of AI applications and addresses the importance of incident detection, common types of incidents, cost of incidents, and benefits that a good incident detection mechanism can provide a roadway network. Finally, it combines the ideas of a decision tree with some AI techniques, and discusses the applications of these AI approaches and the feasibility of computer oriented techniques. Future research concerning on-line incident detection application and evaluation, decision support system for incident handling, and computer learning about incident characteristics are also highlighted.(A)
Abstract