Incident Duration Model on Urban Freeways Based on Discrete Choice Model

Effective traffic incident management requires a full understanding of the various corresponding properties to accurately estimate incident durations and to benefit decision makings of reducing the impact of non-recurring incident relevant congestions. This paper incorporates discrete choice theory in incident duration prediction. Multinomial Logit (MNL) model is employed to forecast the incident duration. All 62941 incident records from Beijing Transportation Management Bureau are used for model establishment and other 10000 records for validation. The average relative error of the model is 27.3%, which shows the reliability of the model is quite satisfactory. Within such a system, the control center operators can estimate the potential impact of a detected incident, and select proper real-time incident management strategies.