Construction Materials · foundry-ml
dataset_concrete_compressive_strength
Concrete mix-design and compressive strength data for civil engineering and materials science machine learning.
What dataset_concrete_compressive_strength is
The dataset records mix-design formulations, curing durations, and compressive strength measurements across 1,030 concrete materials. It was compiled by Zhanzhao Li, Jinyoung Yoon, Rui Zhang, Farshad Rajabipour, Wil V. Srubar III, Ismaila Dabo, and Aleksandra Radlińska, and published in 2022 via the Materials Data Facility and the Foundry platform.
Compressive strength is a foundational mechanical metric in civil engineering, determining whether a structural material meets design criteria. The dataset provides standardised, quantitative mix and test records to support materials-modelling workflows, enabling researchers to explore and evaluate machine learning techniques in concrete science.
How it was made
The data is structured in a tabular schema where each sample represents a specific concrete mixture. Seven continuous input variables define the formulation and test state: proportions of cement, blast furnace slag, fly ash, water, superplasticizer, and coarse aggregate, all measured in kilograms per cubic metre (kg/m³), alongside the curing age measured in days.
The single target variable is the resulting concrete compressive strength, quantified in megapascals (MPa). All 1,030 samples are provided in Parquet format under a single training split.
Key facts
- Total samples
- 1,030 materials
- Data format
- Tabular (Parquet)
- Input features
- 7 (cement, blast furnace slag, fly ash, water, superplasticizer, coarse aggregate, age)
- Target variable
- Concrete compressive strength (MPa)
- Task categories
- Tabular regression, tabular classification
- Release year
- 2022
- DOI
- 10.18126/8k1f-mx77
What you can use it for
- Compressive strength regression
- Fitting regression models to predict mechanical strength in megapascals based on constituent ingredient weights and curing duration.
- Supplementary cementitious material analysis
- Investigating how varying proportions of industrial by-products, specifically fly ash and blast furnace slag, influence concrete strength development.
- Material grade classification
- Developing tabular classification algorithms to group mixtures into discrete strength categories for performance screening.
Explore the data
The schema, real rows and a SQL filter — read in your browser, nothing downloaded.
Explore
Read the real rows without downloading anything
Reading rows…
LiveRead from foundry-ml/dataset_concrete_compressive_strength at the moment you asked. Nothing is cached or stored — every row above came from that request.
How the columns are distributed
Over 1,030 rows of default/train
Age (day)
int- min
- 1
- median
- 28
- mean
- 45.66
- max
- 365
Blast Furnace Slag (component 2)(kg in a m^3 mixture)
float- min
- 0
- median
- 22
- mean
- 73.90
- max
- 359.40
Cement (component 1)(kg in a m^3 mixture)
float- min
- 102
- median
- 272.90
- mean
- 281.17
- max
- 540
Coarse Aggregate (component 6)(kg in a m^3 mixture)
float- min
- 801
- median
- 968
- mean
- 972.92
- max
- 1,145
Concrete compressive strength(MPa, megapascals)
float- min
- 2.33
- median
- 34.45
- mean
- 35.82
- max
- 82.60
Fly Ash (component 3)(kg in a m^3 mixture)
float- min
- 0
- median
- 0
- mean
- 54.19
- max
- 200.10
Superplasticizer (component 5)(kg in a m^3 mixture)
float- min
- 0
- median
- 6.40
- mean
- 6.20
- max
- 32.20
Water (component 4)(kg in a m^3 mixture)
float- min
- 121.80
- median
- 185
- mean
- 181.57
- max
- 247
Strengths and limits
Good at
- Provides standardized physical units (kg/m³, days, MPa) across all formulation and outcome variables.
- Captures diverse binder systems by including supplementary cementitious components such as slag and fly ash alongside conventional cement.
- Offers a dedicated benchmark of 1,030 material formulations specifically curated for machine learning in civil engineering.
Watch out for
- The dataset size is limited to 1,030 instances, which may restrict the training of high-parameter machine learning architectures without regularisation.
- Only a single train partition is specified, requiring practitioners to implement their own cross-validation or testing splits.
Questions
Can I look inside dataset_concrete_compressive_strength without downloading it?
Yes. This page reads its schema, real rows and per-column statistics in the browser and a SQL filter. Nothing is downloaded to your machine.
What licence is dataset_concrete_compressive_strength under?
Its authors declare other. Read the terms on the original page before reusing the data.
What input parameters are recorded for each concrete mix?
Each record includes component quantities in kg/m³ for cement, blast furnace slag, fly ash, water, superplasticizer, and coarse aggregate, as well as the concrete age in days.
What is the primary target measurement?
The target property is concrete compressive strength, reported in megapascals (MPa).
Which machine learning problem types does this dataset support?
It is tagged for both tabular regression and tabular classification workflows.
How to cite it
Li, Zhanzhao and Yoon, Jinyoung and Zhang, Rui and Rajabipour, Farshad and Srubar III, Wil V. and Dabo, Ismaila and Radlińska, Aleksandra. Machine learning in concrete science: applications, challenges, and best practices. Materials Data Facility, 2022. https://doi.org/10.18126/8k1f-mx77
Written from dataset_concrete_compressive_strength's own dataset card. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.