From data to design: Random forest regression model for predicting mechanical properties of alloy steel

TL;DR

This paper demonstrates that Random Forest Regression can accurately predict key mechanical properties of alloy steel from material composition data, offering valuable insights for material design and industrial applications.

Contribution

The study introduces a novel application of Random Forest Regression for predicting alloy steel properties based on composition features, showing high predictive accuracy and potential industrial relevance.

Findings

High R2 scores indicating accurate predictions

Effective use of ensemble learning for material property estimation

Validation through residual plots and learning curves

Abstract

This study investigates the application of Random Forest Regression for predicting mechanical properties of alloy steel-Elongation, Tensile Strength, and Yield Strength-from material composition features including Iron (Fe), Chromium (Cr), Nickel (Ni), Manganese (Mn), Silicon (Si), Copper (Cu), Carbon (C), and deformation percentage during cold rolling. Utilizing a dataset comprising these features, we trained and evaluated the Random Forest model, achieving high predictive performance as evidenced by R2 scores and Mean Squared Errors (MSE). The results demonstrate the model's efficacy in providing accurate predictions, which is validated through various performance metrics including residual plots and learning curves. The findings underscore the potential of ensemble learning techniques in enhancing material property predictions, with implications for industrial applications in material science.