Short-term traffic flow rate forecasting based on identifying similar traffic patterns

摘要: The ability to timely and accurately forecast the evolution of traffic is very important in management control applications. This paper proposes a non-parametric data-driven methodology for short-term forecasting based on identifying similar patterns using an enhanced K-nearest neighbor (K-NN) algorithm. Weighted Euclidean distance, which gives more weight recent measurements, used as similarity measure K-NN. Moreover, winsorization neighbors implemented dampen effects dominant candidates, rank exponent aggregate candidate values. Robustness proposed method demonstrated by implementing it large datasets collected from different regions comparing with advanced time series models, such SARIMA adaptive Kalman Filter models others. It that reduces mean absolute percent error than 25%. In addition, effectiveness K-NN algorithm evaluated multiple steps also its performance tested under data missing research provides strong evidence suggesting approach promising results. Given simplicity, accuracy, robustness approach, can be easily incorporated real-time proactive freeway management.