A Review Study of Incident Detection Algorithms with Performance Index Parameter

This paper proposes a review study of the current incident detection algorithms that can work with two-position traffic data measurements consisting of upstream and downstream. The review study can be considered for the performance comparison by considering the performance index parameter. The performance evaluation is the combination of the results of detection rate, false alarm rate and mean time to detect, enabling the efficiency of all algorithm sequences with single parameter. The review study is a test with simulated data sets according to the research environment requirements reference with a simulation program that is known and recognized internationally as AIMSUN. As the results of the experiment will show the effectiveness of each method even if the performance index parameter is better. On the other hand, the efficiency value in other areas may have an interesting implication that may be appropriate to the needs and acceptance of different errors.