Scalable Infrastructure Intelligence for UIDAI Enrolment & Updates
Project DRAM is a production-grade analytics engine designed to optimize Aadhaar service infrastructure across India. By processing ~5 million records, it identifies demand volatility and provides a strategic roadmap for resource allocation using the Updates-to-Enrolment Ratio (UER).
The system is fully deployed and accessible here:
🔗 Project DRAM Live Intelligence Dashboard
The system follows a modular, three-tier architecture designed for sub-millisecond query performance:
- Data Engine: Robust ingestion layer with validation for multi-source CSV datasets.
- Analysis Layer: Vectorized computation of UER and SciPy-based Z-Score anomaly detection.
- Presentation Layer: Streamlit Cloud interface utilizing 2026 API standards for a warning-free, high-performance user experience.
- ✅ Production Ready: Fully refactored for 2026 Streamlit standards with zero deprecation warnings.
- ✅ Vectorized Performance: Implemented boolean masking for instantaneous regional filtering across national-scale data.
- ✅ Type Safety: Built with strict Python Type Hinting (
Tuple,Optional) to meet enterprise maintainability standards. - ✅ Data Storytelling: Integrated contextual "Logic Expanders" to provide stakeholders with immediate interpretability of saturation metrics.
project-dram/
│
├── main.py # OOP-based Analysis Engine
├── app.py # Modular Dashboard (2026 Production Standard)
├── requirements.txt # Dependency Manifest
├── README.md # System Documentation
│
├── Outputs/ # Strategic Engineering Artifacts
│ ├── final_district_classification.csv
│ ├── anomaly_report.csv
│ └── 5_demographic_insights.png
│
└── Data/ # Input CSV files (UIDAI Dataset)
├── api_data_aadhar_enrolment_*.csv
└── api_data_aadhar_biometric_*.csv
======================================================================= PROJECT DRAM v2.0 - Dynamic Resource Allocation Model UIDAI Hackathon: Unlocking Societal Trends in Aadhaar Data
[STEP 1] Ingesting Multi-Source Data... ✓ Found 12 Data sources | Loaded ~5,000,000 records ✓ Data Integrity Checks Passed (No Nulls in UER columns)
[STEP 5] Classifying Districts into Strategic Zones... ✓ Classification Complete: • RED (Critical Hub): 90 districts • YELLOW (Hybrid): 717 districts
---
## 🚀 Usage & Deployment
### 1. Data Preparation
Place your raw UIDAI source data files into the `Data/` directory:
* `api_data_aadhar_enrolment*.csv`
* `api_data_aadhar_biometric*.csv`
Follow these steps to initialize the analysis engine and launch the dashboard:
# Install required Python dependencies
pip install -r requirements.txt
# Execute the OOP-based Analysis Engine to generate artifacts
python main.py
# Launch the interactive Streamlit Dashboard
streamlit run app.py- Runtime: Python 3.11+
- Core Engine: Pandas, NumPy, SciPy (Z-score calculation)
- Visualizations: Plotly (Sunburst & Interactive Scatter), Seaborn
- UI Architecture: Streamlit (2026 Width-Stretch UI Architecture)
- UER Calculation: Aggregates biometric and demographic updates against total enrolments to determine service demand.
-
Z-Score Anomaly Detection: Utilizes statistical modeling to identify outliers where
$|Z| > 2.5$ , flagging districts with extreme infrastructure stress. - Rule-Based Classification: Implements a dynamic three-tier system (RED, YELLOW, GREEN) to categorize districts for strategic resource allocation.
- Hackathon: UIDAI Innovation Challenge 2026
- Category: Data Analytics & Predictive Modeling
- Shaikh Mohammad Tohid: Lead Software Engineer & Data Analyst -- [shaikhtohid921@gmail.com]
- Solanki Rushikumar: Research & Documentation Lead -- [solankirushi75@gmail.com]
- Email: [shaikhtohid921@gmail.com]
⭐ Star This Repository if you find this project useful for data-driven governance!