THE RESULT
What the data shows
The largest computed metric is 0.9379 for Random forest; the smallest is 0.3814 for Prior baseline. Metric: average_precision (Average precision).
30 MPa is a study-defined threshold, not a code requirement or safety certification.
| model | accuracy | balanced accuracy | macro f1 | roc auc | average precision | brier |
|---|---|---|---|---|---|---|
| Prior baseline | 0.6186 | 0.5 | 0.3822 | 0.5 | 0.3814 | 0.236 |
| Logistic regression | 0.8866 | 0.8669 | 0.8763 | 0.937 | 0.92 | 0.1002 |
| Random forest | 0.8763 | 0.8637 | 0.8675 | 0.9588 | 0.9379 | 0.08336 |
THE METHOD
From source to answer
Binary screening below 30 MPa with composition-grouped evaluation.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
EVALUATION
How the result was checked
Fixed 80/20 split of unique input groups (seed 42); exact input duplicates stay together.
Fixed models; default decision threshold; no holdout tuning.
Target and input features
Target: below_30mpa
Inputs: cement, slag, fly_ash, water, superplasticizer, coarse_aggregate, fine_aggregate, age
THE NEXT DECISION
What follows from the finding
Treat the threshold as an example research screen; project-specific engineering requirements must be supplied separately.
Where the conclusion stops
Laboratory observations do not certify a construction mix. Composition and curing conditions are confounded. Models require independent engineering validation and cannot replace strength testing.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.