: Due to Digitization, there are significant developments and changes occurs in the way people do purchasing, they prefer to buy products online instead of offline. Shopping at just one click of mouse makes everything easy, convenient and at the same time saves a lot of time. Here comes the first challenging task for E-Commerce Companies now a days is to handle bulk returns that is a part of reverse logistics and the second challenge is to make the best possible decisions regarding route and transport in low cost. This paper focuses and gives the solution to above problem by Predicting in advance which product and by whom it is going to be returned in future and after that optimizing the best possible route and transport for returned product is done using AdaBoost model. Computational calculations shows that our framework generates satisfying results in terms of prediction and route optimization. We further shows a comparison between adaboost and regression methods of machine learning by confusion matrix.