Electric Car Price Prediction Using Random Forest Algorithm Comparative Analysis for Türkiye Example and Solution Proposals with Fuzzy Sets

Abstract

Nowadays, due to the adoption of environmentally friendly technologies and the rising fuel prices, consumers are increasingly turning to electric vehicles. However, it is known that one of the most important factors influencing consumers' decisions to purchase electric vehicles is the sales price. In order for the electric vehicle market to grow globally, it is crucial to implement an effective pricing strategy. In this context, the study estimates the sales prices of 96 electric vehicles of various brands, models, and types across different segments using the Random Forest Algorithm. Additionally, while predicting the sales prices, the model was developed based on 10 criteria commonly used in the literature to evaluate the performance of electric vehicles. The results were used to compare the market sales prices and the predicted sales prices of various electric vehicles as part of a case study conducted in Turkey. Furthermore, the relationship between the market and predicted sales prices was analyzed for the four most important quantitative criteria considered in the literature and by consumers when purchasing electric vehicles. In the final section, solution suggestions based on fuzzy logic were provided for situations involving uncertainty and subjectivity. The study is considered to offer guidance to companies in determining the sales prices of electric vehicles.