Georgia Power Large Load Economic Development Pipeline Dashboard

Interactive web dashboard for Georgia Power's large load pipeline data, sourced from quarterly reports filed under Docket 55378 (Q1 2024 – Q2 2026). Built by RMI.

Key Features

Four tab views:

• Pipeline Snapshot — Stacked bar chart of pipeline MW by planning year for a selected quarter
• Pipeline Evolution — Multi-quarter comparison with teal-to-navy color gradient, combining or separating by pipeline stage
• Quarter-over-Quarter Changes — Aggregate metrics (net MW, added/removed projects, schedule delays)
• Snapshot by Vintage — Projects broken down by age bucket (New, 2-3 qtrs, 4+ qtrs)

Filters: Horizontal pill-based filter bars for report quarter, pipeline stage, segment (data center, manufacturing, etc.), vintage, and planning year range. All/None toggles on every filter group.

Load forecast overlays: Selectable dashed reference lines from IRP filings — large load projections and historical system peak demand — plotted alongside the stacked bars in Snapshot and Vintage views.

Table features: Color-coded badges for pipeline stage, segment, and territory. Sortable columns. Client-side search filtering on the projects table.

Data Pipeline

inputs/
  ├── workbooks/             # Raw Excel files per quarter (scraped)
  ├── 2026Q1/                # Q1 2026 manual CSVs (no Excel was filed)
  └── GPC_Load_Forecasts.csv # Reference forecast series
         │
         ▼
scripts/build_dataset.py     # Normalize → combined CSV files
scripts/assign_project_ids.py # Persistent project IDs across quarters
         │
         ▼
outputs/combined/
  ├── pipeline_projects.csv  # Project-level with load columns per year
  ├── pipeline_changes.csv   # QoQ change metrics
  └── pipeline_snapshot.csv  # Aggregated by (quarter, stage, year)
         │
         ▼
scripts/generate_site.py     # Embed JSON into template → index.html
         │
         ▼
index.html                   # Deployable — GitHub Pages ready

Key Scripts

Script	Purpose
scripts/scrape_workbooks.py	Downloads quarterly Excel report files
scripts/build_dataset.py	Parses Excel/CSV → normalized combined datasets
scripts/assign_project_ids.py	Assigns persistent IDs to track projects across quarters
scripts/generate_site.py	Generates index.html with embedded JSON from template
scripts/app.py	Plotly Dash app (live server, same 4-tab layout)
scripts/spot_check.py / spot_check2.py	Validation scripts

File Structure

large-loads-reports/
├── index.html                # Static dashboard (generated)
├── assets/
│   ├── index.template.html   # Source template for index.html
│   └── rmi_logo_horitzontal_no_tagline.svg
├── inputs/
│   ├── 2026Q1/               # Q1 2026 manual CSVs (no Excel was filed)
│   ├── workbooks/            # Quarterly Excel files
│   └── GPC_Load_Forecasts.csv # Reference forecast series
├── outputs/
│   └── combined/             # Normalized CSVs (generated)
├── scripts/                  # All Python scripts
└── README.md

Quick Start

# Regenerate static site after data update
python scripts/generate_site.py

# Or run the live Dash app
python scripts/app.py

Open index.html in any browser — no dependencies needed at runtime.