Electric Vehicle Cost Prediction During the Design Phase Using XGBoost Regressor Method

View at publisher Summarise with VIRAT
Authors: E. Gothai, R. Thamilselvan, P. Natesan, S. Gokul
Cited by: 0
References: 13

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Electric car markets develop well and even reached some 14 million units sold in 2023. Now the EVs have acquired wide popularity as a friendly-to-the-environment transport alternative. Sales of such cars record some significant increase lately. Producing really efficient and reasonably price-effective EVs poses a great challenge for the producers. Some of the biggest hurdles include trying to estimate costs fairly on what it takes to produce during the design and prototype-development phases. Optimum cost predictability in early stages increases profit points and minimizes manufacturing defects in the production process by optimizing design and production workflows. On that basis, during the design phase of electric vehicles, the cost predictability would therefore determine which aspects to optimize in terms of resource allocation and proper decision-making for manufacturers. This paper is an approach toward the prediction of the cost of EV using machine learning. An XGBoost regression model was applied in that setting, mainly to handle missing values and null values, a feature of elegance in the designing stages of early application. That, therefore, would account for a five-module process: Data Preprocessing, Feature Selection and Target Variable Definition, Model Construction, Model Training and Evaluation, and Visualization. Our model is performing better than other algorithms through which it has been trained and evaluated on an extremely comprehensive dataset at a lesser mean absolute error with a performance that had the best prediction. This research study will prove to be a useful tool during the designing stage of the product for the producers of EV so that they can make an estimate of the right costs to be incurred in terms of decisions and optimize the cost. Also, the developed model will further help customers to understand the real cost of EV.

Fields of Study & Subjects

Electric Vehicles and Infrastructure Data Mining and Machine Learning Applications Electric and Hybrid Vehicle Technologies

Publication details

Journal / source
International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
Pages
1–9
Publication date
8 January 2026
Language
EN