Alloy Mechanical Properties · foundry-ml
steel_strength_v1-1
A tabular dataset of elemental compositions and mechanical properties for 312 steel alloys to support alloy design.
What steel_strength_v1-1 is
steel_strength_v1-1 provides chemical compositions and measured mechanical properties for 312 steel alloys. Compiled by Gareth Conduit and distributed through the Materials Data Facility and Foundry-ML, the collection pairs elemental weight concentrations with key mechanical performance metrics.
The dataset addresses the requirement for structured composition–property pairs in engineering and materials science, particularly for machine learning workflows targeted at high-strength alloy discovery and mechanical property estimation.
How it was made
The data is structured as tabular entries where each row corresponds to an individual steel alloy.
The targets comprise three experimentally measured mechanical properties: yield strength in megapascals (MPa), tensile strength in megapascals (MPa), and elongation at fracture as a percentage. All records are provided in a single training split.
Key facts
- Dataset size
- 312 steel alloys
- Target properties
- Yield strength (MPa), tensile strength (MPa), elongation at fracture (%)
- Data format
- Tabular (Parquet)
- Author
- Gareth Conduit
- Release year
- 2022
- Publisher
- Materials Data Facility
- DOI
- 10.18126/524z-vd6m
What you can use it for
- Mechanical Property Regression
- Train supervised regression models to predict yield strength, tensile strength, and elongation from elemental alloy compositions.
- Alloy Design and Screening
- Evaluate hypothetical steel compositions computationally to identify candidates meeting targeted strength and ductility criteria.
- Chromium Content Classification
- Benchmark tabular classification models using the categorical chromium amount attribute.
Explore the data
The schema, real rows, full-text search and a SQL filter — read in your browser, nothing downloaded.
Explore
Read the real rows without downloading anything
Reading rows…
LiveRead from foundry-ml/steel_strength_v1-1 at the moment you asked. Nothing is cached or stored — every row above came from that request.
How the columns are distributed
Over 312 rows of default/train
al
float- min
- 0.01
- median
- 0.05
- mean
- 0.24
- max
- 1.80
c
float- min
- 0
- median
- 0.03
- mean
- 0.10
- max
- 0.43
co
float- min
- 0.01
- median
- 7.08
- mean
- 7.01
- max
- 20.10
cr
float- min
- 0.01
- median
- 9.05
- mean
- 8.04
- max
- 17.50
cr_amount
string_label- high47%
- low32%
- mid21%
elongation
float- min
- 2
- median
- 14.80
- mean
- 14.01
- max
- 35
2.9% empty
formula
string_text- min
- 83
- median
- 105
- mean
- 103.05
- max
- 118
mn
float- min
- 0.01
- median
- 0.01
- mean
- 0.15
- max
- 3
mo
float- min
- 0.02
- median
- 2.21
- mean
- 2.77
- max
- 9.67
n
float- min
- 0
- median
- 0
- mean
- 5.5e-3
- max
- 0.15
nb
float- min
- 0
- median
- 0.01
- mean
- 0.04
- max
- 2.50
ni
float- min
- 0.01
- median
- 8.50
- mean
- 8.18
- max
- 21
si
float- min
- 0.01
- median
- 0.01
- mean
- 0.22
- max
- 4.75
tensile strength
float- min
- 1,019
- median
- 1666.30
- mean
- 1641.65
- max
- 2,570
ti
float- min
- 0
- median
- 0.03
- mean
- 0.31
- max
- 2.50
v
float- min
- 0
- median
- 0.01
- mean
- 0.18
- max
- 4.32
w
float- min
- 0
- median
- 0
- mean
- 0.16
- max
- 9.18
yield strength
float- min
- 1005.90
- median
- 1344.20
- mean
- 1421.00
- max
- 2510.30
Strengths and limits
Good at
- Provides simultaneous measurements of yield strength, tensile strength, and elongation alongside complete compositional profiles.
- Pre-formatted for straightforward integration into tabular regression and classification pipelines.
Watch out for
- The dataset is limited in scale to 312 alloy records.
- All entries reside in a single train split without standard validation or test partitions.
- Thermal histories, heat treatments, and microstructural parameters are not included in the feature set.
Questions
Can I look inside steel_strength_v1-1 without downloading it?
Yes. This page reads its schema, real rows and per-column statistics in the browser, with full-text search and a SQL filter. Nothing is downloaded to your machine.
What licence is steel_strength_v1-1 under?
Its authors declare cc-by-4.0. Read the terms on the original page before reusing the data.
What mechanical properties are recorded in the dataset?
The target variables are yield strength in megapascals (MPa), tensile strength in megapascals (MPa), and percentage elongation at fracture.
Which elemental compositions are detailed?
Weight percentages are provided for carbon (C), manganese (Mn), silicon (Si), chromium (Cr), nickel (Ni), molybdenum (Mo), vanadium (V), nitrogen (N), niobium (Nb), cobalt (Co), tungsten (W), aluminium (Al), and titanium (Ti).
How many records does the dataset contain?
It contains 312 distinct steel alloy samples.
Does the dataset include separate test splits?
No, all 312 records are grouped under a single training split.
How to cite it
Conduit, Gareth (2022). Mechanical properties of some steels. Materials Data Facility. https://doi.org/10.18126/524z-vd6m
Written from steel_strength_v1-1'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.