Estimating the Duration of Motorway Incidents using Machine Learning Approaches

Abstract

Motorway incidents are frequent & varied in nature. The causes & severity of incidents might vary, ranging from road debris to severe traffic collisions. Incident management on motorways is critical for both driver safety & road network operation. The expected duration of an incident is a key parameter in the decision-making process for control room operators. This paper presents a study which compares the ability of different machine learning algorithms to estimate the duration of motorway incidents on Ireland's M50 motorway, using an extensive historical incident database. Results show that the support vector machine has the best performance in most cases, but a different method may need to be used to improve accuracy in some situations. Results highlight difficulties in accurately forecasting incident durations in real time & recommendations are made for improving prediction accuracy.