Multi-Material Tensile Strength Predictor with Machine Learning by Félix Ruiz M.Multi-Material Tensile Strength Predictor with Machine Learning by Félix Ruiz M.

Multi-Material Tensile Strength Predictor with Machine Learning

Félix Ruiz M.

Completed work

Data Scientist

ML Engineer

pandas

Python

TensorFlow

Engineering & Architecture

Advanced Machine Learning system predicting ultimate tensile load for multiple steel materials (A36, S275, S355) using Random Forest algorithm. Trained on 124,639 real tensile test records achieving 90.29% accuracy.

🔹 Multi-Material Support: Handles 3 different steel grades 🔹 High Accuracy: R² = 0.9029 on massive dataset 🔹 Production Ready: Deployed model with API-like interface 🔹 Advanced Features: 8 input parameters including material properties 🔹 Real Data: Based on actual universal testing machine results

Key Capabilities:

Predicts ultimate load without physical testing

Optimizes structural design for cost reduction

Supports engineering decision-making with confidence factors

Handles complex material property relationships

Tech Stack: Python, scikit-learn, Random Forest, pandas, matplotlib, joblib Industry Impact: Reduces testing costs, accelerates material validation, improves design efficiency

Demonstrates scalable ML architecture for engineering applications with practical deployment considerations.

Like this project

Completed work

Posted Jul 28, 2025

ML model predicting ultimate tensile load across 3 steel types (A36, S275, S355). 90.3% accuracy on 124K+ real tensile tests. Production-ready solution.

Likes

Views

Timeline

Jan 3, 2024 - Apr 20, 2024