Automated incident detection algorithms

The high cost of congestion caused by incidents such as accidents, disabled vehicles, construction work and other events that result in a capacity reduction of roads has prompted a growing worldwide interest in developing efficient and effective incident management programs. The success of incident management programs will depend to a large extent on the development of reliable and efficient automated incident detection (AID) algorithms. These software programs analyse data collected from traffic detection devices to determine whether an incident has occurred and alert operators in the traffic control centre to abnormal traffic conditions. This paper provides a background to automated incident detection algorithms and describes some of the most widely used freeway AID models. Common measures of performance and criteria used to evaluate the efficiency and effectiveness of these algorithms are then presented. The traffic detection, software and hardware requirements needed for the development and implementation of these models are also described. A comparative evaluation between a number of AID algorithms is then presented. The paper concludes by highlighting current state-of-the-art techniques in the development of AID algorithms on both freeway and arterial roads and the benefits to be gained from the implementation of quick and reliable incident detection systems.