Predictive analytics for pricing strategy in the automobile industry using machine learning models

Pricing strategy is a critical determinant of success in the highly competitive automobile industry. While traditional models exist, they often lack sophistication and fail to incorporate comprehensive feature engineering. This study addresses these limitations by implementing and evaluating a suite of advanced machine learning algorithms for predicting car prices. We employed Linear Regression, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree, Random Forest, and a Convolutional Neural Network (CNN) on a real-world automotive dataset. A rigorous methodology involving thorough hyperparameter tuning and Explainable AI (XAI) techniques, namely LIME and SHAP, was applied to enhance model performance and interpretability. The models were evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the R-squared (R²) score. Results indicated that the Random Forest model achieved superior predictive accuracy, explaining 92% of the variance in car prices (R² = 0.92), while the CNN excelled at capturing intricate non-linear relationships. Feature importance analysis revealed engine capacity, vehicle age, and year of manufacture as the most significant price determinants. This research demonstrates that leveraging advanced, tuned machine learning models with XAI provides a robust, transparent, and data-driven framework for optimizing pricing strategies, thereby offering significant benefits to automotive industry stakeholders.