RE-source is a modern, data-driven web application designed to optimize material selection for engineering and manufacturing projects. Instead of relying on traditional catalogs, this engine uses a machine learning model to instantly recommend the best materials based on three critical business metrics.
To provide actionable, data-backed material recommendations that help users:
-
Reduce Cost Impact
-
Lower Emissions
-
Maximize Performance Gain
The application follows a simple, yet powerful three-step process:
-
Input: The user provides target values for Cost Impact, Emissions Reduction, and Performance Gain via the web interface.
-
Prediction: The Flask backend routes the user's inputs to a pre-trained XGBoost Regressor Model which analyzes potential - -materials in real-time.
-
Output: The application displays the material recommendations that best fit the user's constraints.
pandas, xgboost, sklearn, joblib, flask, numpy