Exploratory Data Analysis on Halliburton Maintenance Logs

This notebook explores 9 years of maintenance incident data to uncover high-impact issues, downtime trends, and root causes. The analysis helps identify where Halliburton is losing the most machine time—and how better categorization, design changes, or proactive maintenance can improve outcomes.

🔍 Key Insights

1. Downtime Is Highly Concentrated
2. Persistent Axis Problems
3. Digital Readout = Dual Risk
4. Hidden Time Sink

🚀 Next Steps

For interactive exploration and real-time prediction, check out the companion project:

👉 Maintenance Risk Dashboard (Streamlit App) This dashboard allows you to filter data, view key metrics, and simulate machine downtime based on technician notes.

Use it to:

1. Monitor high-risk issues in real time
2. Predict downtime from new technician entries
3. Turn these EDA insights into operational decisions
4. Let me know if you'd like this integrated directly into your current repo structure or documentation.