Traffic Volume Forecast Using Statistical Regression Learning For Annual Average Daily Traffic (aadt) Estimate

Abstract

With The Advent Of Intelligent Transportation Systems (ITS), Short Term Traffic Prediction Is A Major Feature To Develop The Smart Traffic System For The Smart City Framework. It Allows Real Time Data Accumulation, Analysis And Operations In Routing Traffic Through The Cities. From A Systems Perspective, The Traffic Volumes In Large Cities Often Exhibit Spatial And Temporal Coherence Which Can Be Leveraged To Predict The Short Term Traffic Volume. However, The Time Series Prediction Is Challenging Due To The Analysis Of Extremely Large And Complex Data Sets With Dependence On A Multitude Of Parameters Or Features. This Paper Presents A Machine Learning Based Approach Based On The Principal Component Analysis (PCA) And The Back Prop Algorithm To For Sales Forecasting. The Performance Of The Proposed System Is Evaluated In Terms Of The Mean Absolute Percentage Error (MAPE) And The Regression. It Is Shown That The Proposed System Outperforms The Previously Existing System Working On The Benchmark Datasets [1].

Keywords