Vehicle Selection Predictor is a new and important concept in transportation studies. In recent years, prediction model was designed for the prediction of prices of vehicles. In this project, we are trying to build a prediction model for vehicle selection based on its usage. In this project, we consider the problem of vehicle selection for transportation using a multicriteria decision-making approach. The problem includes several conflicting factors which are economic and technological factors. The vehicle has its own specifications/factors like load that it can handle, average of the vehicle, design of vehicle etc. Based on these factors the transport agency should choose a vehicle for a particular delivery. Hence, to predict the selection of vehicle depending on its specifications, the system will be designed by using machine learning algorithms. While designing the model feature selection is a very important aspect. This paper mainly focuses on feature selection and building the predictive model.