Incident Detection Using a Fuzzy-based Neural Network Model

Incidents on the freeway disrupt traffic flow and the cost of delay caused by the incidents is significant. To reduce the impact of an incident a traffic management center needs to quickly detect and remove it from the freeway. In this vein quick and efficient automatic incident detection has been the main goal of the transportation research for many years. Also many algorithms based on loop detector data have been developed and tested for automatic incident detection. However, many of them have a limited success in their overall performance in terms of detection rate, false alarm rate, and the mean time to detect an incident. The objective of this paper is to propose a robust and reliable method for detecting an incident on the freeway using a fuzzy based neural network model, Fuzzy ARTMAP which is a supervised, self-organizing system claimed to be more powerful than many expert systems, genetic algorithms, or other neural network models like Backpropagation. The experiments have been done with simulated data, and the results show that Fuzzy ARTMAP has the potential for the application of automatic incident detection in the real world, where a large number of incident data is not always available for training.