Data-driven formulation of steel fiber pull-out force in cementitious composites using genetic programming.

Area of Science:

• Materials Science
• Civil Engineering
• Computational Mechanics

Background:

• Steel fiber reinforced cementitious composites are crucial in construction.
• Predicting steel fiber pull-out force is vital for structural integrity.
• Existing methods often rely on extensive experimental testing.

Purpose of the Study:

• To develop a data-driven model for predicting steel fiber pull-out force.
• To formulate an accurate and interpretable mathematical expression for pull-out force.
• To reduce the need for time-consuming and expensive experimental procedures.

Main Methods:

• Utilized a comprehensive dataset of 437 experimental results.
• Employed gene expression programming (GEP), a variant of genetic programming.
• Performed k-fold cross-validation and SHAP sensitivity analysis for model validation.

Main Results:

• Achieved a high prediction accuracy with an R² of 0.93.
• Identified fiber tensile strength and embedment length as key influencing factors.
• Proposed a novel, interpretable mathematical formula for pull-out force.

Conclusions:

• The GEP model provides a robust and accurate method for predicting pull-out force.
• The derived formula offers an efficient tool for engineers and researchers.
• This data-driven approach enhances performance prediction and formulation in composite materials.