Travel Time Prediction Using a Seasonal Autoregressive Integrated Moving Average Time Series Model

Travel time estimation and prediction form an integral part of any advanced traveler information system. This paper presents a univariate time series based approach to predicting future travel times using historical travel time data. This model relies strictly on point detection data. Empirical testing of the model is performed using ITS data obtained from video detection systems in Atlanta, Georgia. The results of the model choice, validation and testing are reported and conclusions interpreting the findings in the model development process are provided