Machine Failure Prediction September 2023

Conducted an extensive investigation of a Milling machine to proactively identify and prevent machine failures, focusing on enhancing operational reliability.

Technical stack Used in the Project -

• Analyzed the dataset comprising 10,000 data points, featuring 14 distinct machine-specific features, to identify potential failures and optimize machine performance.
• Leveraged a diverse set of machine learning algorithms, culminating in an impressive 97\% recall rate for predicting machine failures. This achievement was made possible through the strategic application of logistic regression and addressing class imbalance with the SMOTE approach.
• Skillfully managed in Azure Designer, within Azure Machine Learning Workspaces, showcasing the capacity to construct machine learning pipelines for in-depth data analysis and predictive modeling.

The Github code is here