An end-to-end operational analytics project simulating the real-world setup of the HCL–Foxconn Semiconductor OSAT Facility in Uttar Pradesh, India. This project includes production analytics, equipment downtime tracking, workforce optimization, absenteeism prediction, SQL-based supply chain analysis, and Power BI dashboarding.
osat-data-analyst-project/
│
├── notebooks/
│ ├── cleaning_exploration.ipynb
│ ├── production_analysis.ipynb
│ ├── equipment_downtime.ipynb
│ ├── workforce_dashboard.ipynb
│ └── predictive_modeling.ipynb
│
├── dashboards/
│ ├── powerbi_mockup.png
│ └── powerbi_notes.md
│
├── sql/
│ └── supply_chain_delay_queries.sql
│
├── data/
│ ├── production_data.csv
│ ├── equipment_logs.csv
│ ├── workforce_schedule.csv
│ └── supply_chain_data.csv
│
├── README.md
└── requirements.txt
- Analyze chip production efficiency and yield
- Investigate equipment downtimes and maintenance patterns
- Visualize workforce productivity and shift-level absenteeism
- Build predictive models for yield and absenteeism risk
- Write SQL queries to identify supply chain bottlenecks
- Design a complete Power BI mockup dashboard for factory operations
| Dataset | Description |
|---|---|
production_data.csv |
Daily production: chips, lines, defects, shifts |
equipment_logs.csv |
Downtime by machine, error logs, MTTR |
workforce_schedule.csv |
Daily attendance by department & shift |
supply_chain_data.csv |
Regional supply delays and vendor issues |
- ✅ Yield Forecasting using Random Forest Regressor
- ✅ Absenteeism Classification with Logistic Regression
- 🔍 Metrics: RMSE, R², Accuracy, Confusion Matrix, ROC Curve
- KPIs: Yield %, Defect rate, Absenteeism rate, MTTR
- Visuals: Time-series trends, heatmaps, department performance
- Screenshot:
dashboards/powerbi_mockup.png
- Python (Pandas, Seaborn, Scikit-learn, Matplotlib)
- SQL (for supply chain insights)
- Power BI (Mock dashboard)
- Jupyter Notebooks
Harsh Sonkar Data Analyst | AWS Data Engineer | Operational Intelligence 🔗 LinkedIn 🌐 GitHub
MIT License