Datasets
Synthetic dataset
Scrap Blend Optimization Dataset
Scrap stream assays and blend ratios paired with resulting chemistry, impurity risk, and yield for blend-planning tasks.
Scrap & recycled metalsScrap & Secondary Metals IntelligenceScrap optimizationCC-BY-4.0
30k
rows
29
features
3.4k
downloads
This dataset is intended for research, education, and benchmarking. Industrial deployment requires validation with plant-specific data.
Schema
| Field | Type | Description |
|---|---|---|
| stream_assays | float[] | Scrap stream assays |
| blend_ratios | float[] | Blend fractions |
| impurity_risk | float | Impurity risk (label) |
| yield | float · % | Yield (label) |
Sample preview
| blend_id | stream_A | stream_B | imp_Cu | impurity_risk | yield |
|---|---|---|---|---|---|
| b-6601 | 0.6 | 0.4 | 0.18 | 0.22 | 96.1 |
| b-6602 | 0.5 | 0.5 | 0.26 | 0.41 | 94.3 |
| b-6603 | 0.7 | 0.3 | 0.14 | 0.17 | 96.8 |
| b-6604 | 0.4 | 0.6 | 0.32 | 0.55 | 92.7 |
Data dictionary
- impurity_riskPredicted impurity-breach risk for the blend (label)
- yieldMetal yield of the blend (label, %)
Discussion
- r.Question
r.okafor
Digital product lead · 4d
Is the synthetic generator's noise model documented anywhere? Want to make sure augmentation is comparable to what we add on our side.
9 - maComment
matereal
Maintainer · 4d
Generator notes are in the linked research summary. Happy to expand the data dictionary if any fields are unclear.
6
Open by design. Private by principle.
Build on open metal intelligence, keep your plant advantage private
Explore open reference models and datasets, or start a private technical discussion. Open industry knowledge and private customer advantage are kept separate.