A Machine Learning Framework for Electrostatic Discharge Accident Prevention

This work presents a semester long project focused on the prevention of Electrostatic Discharge (ESD) accidents through data driven modeling techniques. The primary objective of the project was to analyze key environmental and operational factors contributing to ESD incidents and develop predictive models to better understand and mitigate associated risks.

The dataset used in this study included variables such as humidity levels, equipment usage, grounding conditions, and human handling practices. A Random Forest model was implemented to classify and predict the likelihood of ESD related incidents under varying conditions. This approach enabled us to identify the most influential features contributing to accident risk and provided insights into how different factors interact within ESD sensitive environments.

Significant emphasis was placed on data preprocessing, including data cleaning, feature selection, and preparation to ensure model accuracy and reliability. Based on the results, several preventive recommendations were proposed, such as maintaining optimal environmental conditions, improving grounding protocols, and reinforcing safe handling practices.

Overall, this project demonstrates the effectiveness of machine learning based data modeling in enhancing safety measures and proactively preventing ESD accidents. The findings highlight the potential for integrating predictive analytics into real-world industrial safety systems to improve reliability and reduce risk.

• Intermediate
• AI/ML