System architecture of a decision support system for freeway incident management in Republic of Korea

Abstract

Traffic Management Centers play a vital role in efficient functioning of freeway network in the post-incident scenario. As per present practice in Korea, the traffic managers use a heuristic approach for incident analysis based on their experience of similar scenarios. However, this approach induces uncertainty thereby reducing the overall effectiveness of the subsequent incident management and rescue operations. This paper proposes a decision support system to account for these shortcomings. We name our system as ‘FIAS’ – Freeway Incident Analysis System. The novel idea presented in this paper is the use of historical, real-time and spatial data simultaneously to forecast post-incident traffic flows on a microscopic simulation platform, Cellular Automata. FIAS incorporate two additional rules in the conventional model to depict more realistic incident flow characteristics. This paper focuses on the system architecture of the model and tests its performance by comparing its predicted values with a real incident data. The evaluation results confirm the validity of FIAS as it can model the time dependent microstructure of traffic flows with significant accuracy.