Datadory notebook
Construction Materials Data Guide: Every Dataset Worth Knowing in 2026
1,744 datasets. Pick your catch. Every guide here is built on what the catalog can actually prove.
1,744 datasets. Pick your catch.
The construction materials data landscape in 2026
Construction materials data splits cleanly into four clusters, and knowing which cluster you need is most of the battle.
Laboratory and materials-science records live in open repositories. Zenodo hosts the 27-Year Multi-Producer European Cement Characterization Dataset (476 samples from 23 anonymized European producers, measured between 1997 and 2024) and the FutuRaM building composition raw data; figshare carries BMAT's 22.09 million facade-material footprints; and the UCI Machine Learning Repository supplies the classic 1,030-row Concrete Compressive Strength Dataset.
Asset-level coverage is unusually strong for this industry. Global Energy Monitor's Global Cement and Concrete Tracker lists 3,884 plants across 171 countries and areas, and the Spatial Finance Initiative's commercial delivery terms database on Dryad links those plants to upstream mine suppliers, an attribution no other cement asset database provides.
What are the core construction materials datasets?
Government and institutional statistics
Japan's e-Stat tabulates the 2023 Housing and Land Survey (October 2023, conducted roughly every five years) into 38 database tables plus 36 downloadable files covering construction material of dwellings across major metropolitan areas. API access needs a free registered application ID, and e-Stat's terms permit commercial use with source citation.
Open science and laboratory repositories
The 27-Year Multi-Producer European Cement Characterization Dataset is uniquely longitudinal: 476 cement samples (311 CEM I and 165 other types) from 23 anonymized producers, with 50 variables spanning chemistry, fineness, setting behaviour and 2/7/28-day strength, in 194.3 kB across four CSV files. Producer identities are deliberately anonymized and must not be re-identified.
The Concrete Compressive Strength Dataset from the UCI Machine Learning Repository remains the standard machine-learning benchmark for mix-design modeling: 1,030 laboratory mix records by 9 attributes (cement, blast furnace slag, fly ash, water, superplasticizer and aggregate quantities in kg per m3 plus age in days, mapped to strength in MPa), donated in August 2007 and downloadable as a ~34 KB ZIP under commercial delivery terms 4.0 with the Yeh (1998) citation requirement.
FutuRaM's Building Composition and Material Intensity Raw Data brings 7,346 material-intensity entries in kg/m2 across 28 predominantly EU/EEA countries and construction years 1859-2012, plus elemental composition of metals in the EU construction sector; version 2 landed in May 2026.
BMAT (the Global Building Facade Material Dataset) labels 22.09 million individual building footprints across 73 major world cities with facade materials (brick, concrete, glass, metal, stone, stucco, tile, wood, other) inferred from 147 million multi-temporal street-view images collected 2007-2025. It ships as 73 city GeoPackages totalling about 20.6 GB, with version 2 posted 2026-06-09.
Building Material Stocks and Embodied Carbon (Springer Nature / Scientific Data) quantifies in-use stocks of 11 materials and cradle-to-gate embodied carbon for 38 cities in China's Yangtze River Delta and Greater Bay Area from 2000 to 2020, at city and 500x500 m grid scale, in 26 xlsx tables totalling 132,931,068 records.
Production and emissions time series
Global CO2 Emissions from Cement Production (GCP-CEM) holds the deepest history available anywhere in this industry: annual cement and clinker production and computed process CO2 for 214 countries from 1880 to 2025 (clinker from 1900), with every data point attributed to its source, in about 2.5 MB across nine files. Nine countries get detailed national workbooks, including China back to 1949 with monthly NBS series. The May 2026 release is version 260526.
Asset-level commercial intelligence
Price feeds and API-native sources
How is construction materials data used?
Eight use cases recur across this slice, and each maps to named datasets:
Notice the pairing pattern: nearly every use case combines one broad structural dataset with one granular one. The lab benchmark pairs with the longitudinal validation set; the plant tracker pairs with the supplier database; the annual emissions series pairs with the monthly production collation.
Who uses construction materials data?
Four user groups dominate demand for this slice, and each searches differently.
Market researchers and strategy teams search for a 'cement plant list with capacity' or 'cement production data by country' ahead of market-entry or M&A work. Their stack is the Global Cement and Concrete Tracker's 3,884 plants joined to the Spatial Finance Initiative's ownership and supplier identifiers, sized against GCP-CEM's 214-country production histories. The market researchers playbook walks that screen end to end.
Sustainability analysts search 'embodied carbon data construction materials' and combine NIST's openLCA templates, the 500x500 m gridded embodied carbon tables for 38 Chinese cities, and GCP-CEM process CO2 computed from clinker production.
How do you build a construction materials data stack?
A working construction materials stack assembles in six steps, each grounded in a dataset above:
For demand-side extension beyond this industry, two related records in the slice help: the US Census New Residential Construction dataset (housing starts, completions and sales) from the real estate industries, and the US BLS Public Data API (PPI, CES and OES series) for producer-price context. Both are cataloged as related datasets for Construction Materials, bringing the pooled count to 19.
What the numbers say
These are the eight figures an analyst is most likely to quote from this catalog, each traceable to a specific field in the 17 primary records below:
| Metric | Value | Source of the number |
|---|---|---|
| Datasets cataloged for this industry | 19 (17 primary + 2 related) | Datadory industry slice |
| Datasets scoring 9/10 quality | 4 | quality_score fields |
| Largest single dataset | ~20.6 GB (BMAT, 73 city GeoPackages) | estimated_size |
| Deepest time series | 1880-2025, 214 countries (GCP-CEM) | coverage.temporal |
| Asset-level coverage | 3,884 cement plants, 171 countries (GEM tracker) | coverage.geographic |
Want rows instead of a pitch? Name the datasets.
API, files, or your warehouse. Daily, weekly, or hourly.
Get a sampleQuestions worth asking
Where can I find free cement production data by country?
Global CO2 Emissions from Cement Production (GCP-CEM) publishes annual cement and clinker production for 214 countries from 1880 to 2025 as commercial delivery terms 4.0 CSVs on Zenodo, with every data point source-attributed. For monthly frequency, Monthly Global Cement Production Data (CICERO) links 70 series across 24 countries to live statistical sources.
Is there a dataset of building facade materials by city?
BMAT, the Global Building Facade Material Dataset, labels 22.09 million individual building footprints across 73 major world cities with facade materials such as brick, concrete, glass, metal, stone and wood, inferred from 147 million street-view images. It ships as 73 city GeoPackages totalling about 20.6 GB under commercial delivery terms 4.0 on figshare.
What dataset can I use to predict concrete compressive strength?
The Concrete Compressive Strength Dataset from the UCI Machine Learning Repository gives 1,030 mix records: cement, blast furnace slag, fly ash, water, superplasticizer and aggregate quantities in kg per m3 plus age in days, mapped to strength in MPa under commercial delivery terms 4.0. The 27-Year Multi-Producer European Cement Characterization Dataset adds 476 industrial samples with 50 variables.
Which sources track cement plant locations and capacity worldwide?
Two complementary asset-level databases exist. Global Energy Monitor's Global Cement and Concrete Tracker covers 3,884 plants across 171 countries with capacity, kilns and green-cement technology, released annually under commercial delivery terms 4.0. The Global Database of Cement Production Assets and Upstream Suppliers (Spatial Finance Initiative, Oxford) adds geolocated plants with ownership identifiers and upstream mine suppliers under commercial delivery terms.
How do I get monthly construction material prices by state?
Malaysia's CIDB N3C Building Material Price API returns documented JSON with monthly average prices for specifications like cement, aggregates and bricks, filterable by state and date range using a Bearer token from an N3C account. For India, the InfraLens-based Apify actor normalizes cement, TMT steel, sand, RMC and brick prices across cities, with InfraLens covering 50-plus cities.
Where do I find embodied carbon data for construction materials?
Three sources cover it. NIST publishes openLCA life-cycle assessment models for cement, concrete and gypsum board bridged to Federal LCA Commons databases. Building Material Stocks and Embodied Carbon quantifies cradle-to-gate emissions for 38 Chinese cities at city and 500x500 m grid scale from 2000 to 2020, and GCP-CEM computes process CO2 from clinker production.