A NEW TRAFFIC SPEED FORECASTING METHOD BASED ON BI-PATTERN RECOGNITION

Short-term traffic forecasting has played a key role in supporting the need of proactive and dynamic traffic control system. K-nearest neighbor (KNN) nonparametric regression models have been widely used in traffic prediction. KNN models give predictions based on the future state of traffic speed that is completely determined by the current state, but with no dependence on the past sequences of traffic speed that produced the current state. In fact, traffic speed is not completely random in nature, and some patterns repeat in the traffic stream. In this paper, we proposed a methodology called bi-pattern recognition KNN model (BKNN) which uses pattern recognition technique twice in the searching process to predict the future traffic state. Then the proposed BKNN model is applied to predict one day real traffic speed series of two sites, which are located near the North 2nd and 3rd Ring Road in Beijing, respectively. With the optimal neighbor and pattern size, the BKNN model provides good predictions. Moreover, in comparison with the KNN model, PKNN model (a modified model based on KNN), seasonal autoregressive integrated moving average (SARIMA) and the artificial neural networks (ANN), the BKNN model appears to be the most promising and robust of the five models to provide better short-term traffic prediction.

[1]  Michael J Demetsky,et al.  TRAFFIC FLOW FORECASTING: COMPARISON OF MODELING APPROACHES , 1997 .

[2]  Melvyn D Cheslow,et al.  AN INITIAL EVALUATION OF ALTERNATIVE INTELLIGENT VEHICLE HIGHWAY SYSTEMS ARCHITECTURES , 1992 .

[3]  Laurence R. Rilett,et al.  Travel Time Prediction by Advanced Neural Network , 2002 .

[4]  Yves Lechevallier,et al.  SHORT TERM TRAFFIC FORECASTING USING NEURONAL NETWORK , 1994 .

[5]  Lawrence W. Lan,et al.  Investigation of temporal freeway traffic patterns in reconstructed state spaces , 2008 .

[6]  N. Altman.  An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression , 1992 .

[7]  Xuewei Li,et al.  Chaotic analysis of traffic time series , 2005 .

[8]  Young-Ihn Lee,et al.  Short-term travel speed prediction models in car navigation systems , 2006 .

[9]  Angshuman Guin,et al.  Travel Time Prediction Using a Seasonal Autoregressive Integrated Moving Average Time Series Model , 2006, 2006 IEEE Intelligent Transportation Systems Conference.

[10]  Billy M. Williams,et al.  Comparison of parametric and nonparametric models for traffic flow forecasting , 2002 .

[11]  Balaji Rajagopalan,et al.  Statistical Nonparametric Model for Natural Salt Estimation , 2005 .

[12]  B. G. Ratcliffe,et al.  Short term traffic forecasting using time series methods , 1988 .

[13]  W. Lam,et al.  Comparison of Four Modeling Techniques for Short-Term AADT Forecasting in Hong Kong , 2003 .

[14]  Stephen D. Clark,et al.  Traffic Prediction Using Multivariate Nonparametric Regression , 2003 .

[15]  Pengjian Shang,et al.  Nonlinear analysis of traffic time series at different temporal scales , 2006 .