Incident detection in signalized urban roads based on genetic algorithm and support vector machine

Detecting incidents is the important step for efficient incident management which aims to minimize the impact of non-recurrent congestion. A little research has been performed to automatically detect incidents in urban arterials. This paper provides incident detection system based on Support Vector Machine (SVM) in urban traffic networks using “Original urban network scenario” and “Freeway like scenario”; moreover, for the best performance of the classifier we introduce an optimization using genetic algorithm (GA). The results show that Genetic Algorithm Support Vector Machine (GA-SVM) has a high classification accuracy and high detection performance with regard to the detection rate and false alarm. A comparative study confirms that the effectiveness of GA-SVM mainly for signalized urban roads scenario.