Inzinion/energy-industry-facilities-composite

Energy & Industry Facilities Composite

Power plants and industrial facilities worldwide in one consistent schema, with coordinates, capacity, owner, status, commissioning year and annual CO2 where available - regulator-reported and modelled values separated by a co2_basis flag.

Canonical record: doi.org/10.5281/zenodo.22172583 · Publisher: Inzonex

Load

from datasets import load_dataset

ds = load_dataset("Inzinion/energy-industry-facilities-composite", split="train")
print(ds)

Or straight into pandas:

import pandas as pd

df = pd.read_csv("https://huggingface.co/datasets/Inzinion/energy-industry-facilities-composite/resolve/main/energy_industry_facilities.csv")

73,386 facilities in one schema — 42,853 power plants + 30,533 industrial facilities · 211 countries · CC BY 4.0

Power-sector and industrial facility data live in different datasets with different schemas. This composite puts both in one consistent table: facility_class (power | industry), sector (primary fuel for power, subsector for industry), country/ISO3, coordinates, capacity with unit, owner, lifecycle status and commissioning year (power layer), and annual CO₂ where available — 35,426 facilities with CO₂, of which 1,239 regulator-reported (EU ETS / US EPA GHGRP) and 34,189 modelled (Climate TRACE), separated by the co2_basis flag. The arithmetic sum of the per-record CO₂ values is 19.86 Gt — a sum of source records, not a deduplicated global annual total (see Honest limits: layers are not cross-matched and reporting years/boundaries differ between sources).

Files

File	Contents
data/energy_industry_facilities.csv	73,386 facilities × 18 columns
data/country_summary.csv	Per-country totals: facility counts by class, total and reported CO₂
data_dictionary.csv	Field-level documentation

Positioning

This is a foundation layer, not an analysis: its value is a consistent cross-sector schema with provenance kept per record (facility_id preserves the source identifier — WRI GPPD / GEM / Climate TRACE — and co2_source/co2_basis keep estimates and regulatory figures apart). Duplicate facilities across the two layers (e.g. an on-site power plant at an industrial complex) are not cross-matched in this version; identifier crosswalks between GEM, Climate TRACE, EIA and EU ETS registries are the planned next step and, for researchers, the more valuable artifact.

Honest limits

• Industry-layer CO₂ is entirely modelled (Climate TRACE estimates); reported values exist mainly in the power layer (EU ETS, US EPA GHGRP).
• Owner coverage is 50% overall: 74.5% in the power layer, 15.1% in the industry layer (several industry subsectors carry no owner at all).
• The 19.86 Gt CO₂ figure is an arithmetic sum of per-record values across both layers without cross-layer deduplication or year/boundary harmonisation; do not quote it as a global annual emissions total.
• Industry capacity units are heterogeneous (capacity_unit per row); do not sum capacity across subsectors.
• Facility detection and sector tags inherit each source's methodology; this is a merge with normalisation, not new measurement.

Sources and attribution

WRI Global Power Plant Database · Global Energy Monitor · Climate TRACE · EU ETS Union Registry / EUTL · US EPA GHGRP. Derived and redistributed under the sources' open licenses; see datapackage.json.

Versions

v1.1.0 (2026-08-30): negative numeric values (longitudes, latitudes, generation) are now written as plain numbers — v1.0.0 escaped them with a leading apostrophe as an over-broad spreadsheet-formula guard, which broke numeric parsing; owner-coverage figures corrected (v1.0.0 claimed "95%+ in the power layer"; the computed figures are 74.5% power / 15.1% industry); CO₂ total rephrased as an arithmetic sum of source records; croissant.json upgraded to conformant MLCommons Croissant 1.0.

Cite

See CITATION.cff. Published by Inzonex — engineering data for high-temperature plant equipment.

Citation

Canonical DOI (always resolves to the newest version):

Aheiev, D. (Inzonex) (2026). Energy & Industry Facilities Composite [Data set]. Zenodo. https://doi.org/10.5281/zenodo.22172583

@dataset{inzonex_energy_industry_facilities_composite,
  author    = {Aheiev, Dmytro},
  title     = {Energy & Industry Facilities Composite},
  year      = {2026},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.22172583},
  url       = {https://doi.org/10.5281/zenodo.22172583}
}