An interactive and data-driven dashboard built using Shiny for Python to analyze the U.S. electricity landscape across Operable, Proposed, and Retired power plants.
This dashboard provides insights on:
- Power plant distribution across states
- Technology-wise capacity trends
- Retirements and new additions
- Plant generation patterns
- Suitability scoring for future energy projects
- State-level maps, treemaps, sunburst charts & Sankey diagrams
The tool helps energy stakeholders and analysts understand the capacity mix, growth patterns, and strategic opportunities across the United States.
In this multi-member capstone, my primary work included:
- Designing and building the modular Shiny for Python UI
- Implementing multi-state dropdown filters and dynamic sheet switching
- Integrating Plotly charts for capacity, status, and distribution comparisons
- Cleaning, merging, and preparing the Operable, Proposed & Retired datasets
- Creating combined analysis logic and capacity calculations
- Deploying the dashboard on AWS EC2 with a reproducible environment
- Creating visualizations for technology distribution, capacity trends & recommendation scoring
- Total utilities, plants, and generators
- Capacity by technology (Nameplate, Summer, Winter)
- Sector vs Status grouped bars
- Uprates, derates, repowers
- Technology distribution dot plot
- Technology-year summary table
- Summary metrics
- Sector vs status visual
- Effective vs current year scatter
- Planned retirement bubble chart
- Summary metrics
- Capacity + generator retirements over time
- Sector vs status visual
- U.S. map of all plants
- Capacity treemaps
- Technology distribution
- Net generation map
- Fuel-type Sankey diagram
- Recommendation scoring modal
- Summary tables for energy mix
Used to evaluate whether a state is suitable for future power project investments:
Score = 0.5 × (Proposed Plants / (Retired Plants + 1))
+ 0.5 × (Proposed Capacity / (Retired Capacity + 1))
Categories:
| Score | Recommendation |
|---|---|
| ≥ 2.0 | Highly Favorable |
| 1.0 – 1.99 | Favorable |
| 0.5 – 0.99 | Neutral |
| < 0.5 | Unfavorable |
/app
main.py
/data
operable_sample.csv
/dashboard_images
Operable.png
Proposed.png
Retired.png
Combined_Analysis.png
README.md
Run the following command:
pip install pandas numpy plotly shiny dash openpyxl
shiny run --port 8000 app/main.py
The dashboard will open in your browser at:
The full cleaned datasets used in the capstone are large and therefore not included in this repository.
A 20-row sample (operable_sample.csv) is provided to illustrate structure and preprocessing steps.
This project was originally developed as a team capstone at George Mason University.
This repository reflects my personal contributions, cleaned structure, and documentation for portfolio purposes.
Email: sriyareddy696@gmail.com
Feel free to reach out for any questions or feedback.