Development of a Travel Time Prediction Method using Data obtained from the Toll Collection System

Abstract

Recently, travel time information is very important in traffic control systems. At present, travel time information is calculated by sensor data such as traffic counter data. However, this method can not consider a change of road traffic conditions in the future. Therefore, this paper proposes a travel time prediction method considering future traffic conditions by using pattern matching. The prediction method adopts pattern matching with travel time pattern. The prediction method uses a best matching pattern to be extracted from accumulated travel-time-patterns. Furthermore, this method is improved upon for a few accumulated patterns. Moreover, to consider applying ETC, proposed system uses the toll collection system data. For confirmation of prediction accuracy, this paper verifies the prediction method by using actual data. As a result, it was confirmed that the travel time prediction value obtained from the toll collection system was the allowance level.