Interconnection Queue Tracker

Search generator interconnection queue projects across US ISOs/RTOs: the public list of power projects waiting to connect to the grid.

Open the live dashboard — explore the interface without setting up a database or downloading the feeds.

Interface captured September 4, 2026. Market coverage and totals reflect the latest available source reports.

Engineering overview

The Next.js/TypeScript interface reads project data from PostgreSQL. Python workers normalize the public ISO reports, and GitHub Actions schedules the daily ingestion. The ERCOT large-load panel uses a separate snapshot from the Kardashev data service.

For a quick code walkthrough, start with the application routes, UI components, and the ingestion service. Setup and deployment instructions follow below.

Supported feed integrations: ERCOT, MISO, PJM, CAISO, SPP, NYISO, ISO-NE, each from that market’s public queue report, fetched directly from each ISO’s public data source. Dashboard coverage depends on the latest successful ingestion; a supported integration does not guarantee currently available rows.

---

Project layout

interconnection-queue-tracker/
├── web/                  # Next.js app (deploy this on Railway)
│   ├── app/              # routes, layout, metadata
│   ├── components/       # UI
│   ├── data/             # curated fallback JSON
│   └── lib/              # DB access, analytics, markets
├── services/fetcher/     # Python job: loads ISO feeds into Postgres
├── scripts/              # local fetch helper
├── docker-compose.yml
├── Dockerfile.web
└── .github/workflows/    # daily fetch CI

---

Quick start

docker compose up -d postgres

cd web && cp .env.local.example .env.local && npm install && npm run dev

# Load all ISO queues (~2-5 min)
docker compose --profile fetch run --rm fetcher

Open http://localhost:3000

Fetch specific ISOs only

FETCH_MARKETS=PJM,CAISO docker compose --profile fetch run --rm fetcher

Or locally without Docker:

./scripts/fetch-queues.sh

---

Stack

Layer	Tech
UI	Next.js 16, Tailwind v4
Data	Postgres (queue_projects, queue_market_snapshots)
Fetcher	Python (direct ISO feeds)

The web app reads Postgres directly (no separate API service). The ERCOT large-load panel is the exception: it fetches a pre-built snapshot from kardashev-data over HTTP.

---

Supported markets

ISO/RTO	Source
ERCOT	GIS Report
MISO	GI interactive queue
PJM	Public interconnection queue export
CAISO	Public queue report
SPP	Generation interconnection summary
NYISO	Interconnection queue
ISO-NE	Interconnection queue

Not yet: IESO, AESO

---

Environment

Web (web/.env.local)

Variable	Example	Purpose
DATABASE_URL	postgres://queue:queue@127.0.0.1:5434/queue	Live queue data
CURATED_PATH	./data/curated.json	Fallback signals
NEXT_PUBLIC_SITE_URL	https://your-app.up.railway.app	OG metadata
KARDASHEV_DATA_URL	https://data.kardashevlabs.org	Optional. Source for the ERCOT large-load panel; defaults to the hosted kardashev-data service if unset.

Fetcher (services/fetcher)

Variable	Purpose
DATABASE_URL	Required. Postgres connection string.
FETCH_MARKETS	Optional. Comma-separated subset of markets to fetch (default: all seven).
PJM_QUEUE_SUBSCRIPTION_KEY	Required for PJM. Public subscription key used by PJM's interconnection queue web app (not a member API key). Without it, PJM is skipped.

---

Production

Railway (web only)

1. Create a Railway project with PostgreSQL + a web service.
2. Set the web service root directory to web.
3. Link Postgres → injects DATABASE_URL.
4. Set NEXT_PUBLIC_SITE_URL to your Railway URL.
5. Add GitHub secrets DATABASE_URL (Railway Postgres public URL) and PJM_QUEUE_SUBSCRIPTION_KEY for the daily fetch workflow (without the PJM key, PJM is skipped).
6. Run the fetcher once before launch (Actions → “Fetch ISO queues” → Run workflow).

Docker

docker compose up -d postgres web
docker compose --profile fetch run --rm fetcher

Scheduled fetch (daily)

GitHub Actions runs at 06:00 UTC; needs repo secret DATABASE_URL.

Docker cron (self-hosted):

docker compose --profile cron up -d fetcher-cron

---

Launch checklist

• Postgres running with DATABASE_URL set on web
• Fetcher has populated queue_projects at least once
• NEXT_PUBLIC_SITE_URL matches your public URL
• npm run lint and npm run build pass in web/