Table 1 Existing literature about ML models utilization in concrete properties prediction.
Ref | Type of Concrete | Models applied | No. of Inputs | Dataset | Performance Metrics | Best Model | ||
|---|---|---|---|---|---|---|---|---|
R2 | RMSE | MAE | ||||||
Fly ash Based concrete | ANN PSO ANN—ICA | 0.838 0.877 | - | - | ANN-PSO | |||
Waste steel slag | ANN MLR M5P FQ | 6 | 338 | 0.986 0.868 0.912 0.869 | 1.34 4.14 3.39 4.10 | 1.05 3.21 2.45 3.18 | ANN | |
Waste Tire Rubber | NLR MEP ANN MARS | 8 | 135 | 0.944 0.951 0.979 0.982 | 6.2 5.7 3.7 3.5 | MARS | ||
HSC | GEP | 10 | 32 | Error < 14% | ||||
SCC | M5P FQ MLR LR | 9 | 436 | 0.91 0.87 0.42 0.67 | 4.22 5.22 12.42 7.95 | M5P | ||
UHPFRC | LR ANN M5P FQ | 10 | 192 | 0.81 0.93 0.88 0.86 | 8.72 4.97 6.84 8.50 | ANN | ||
Rubberized concrete | GEP ANN Bagging | 4 | 324 | 0.982 0.984 0.968 | 0.918 0.867 1.211 | 0.730 0.621 0.928 | ANN | |
POFA concrete | ANN ANNX PSO GA | 6 | 249 | 0.989 0.977 0.934 0.923 | 0.033 0.049 0.082 0.087 | 0.023 0.023 0.057 0.066 | ANN and ANNX | |
UHPC | Hybrid XGB LGB | 13 | 317 | 0.963 0.946 0.943 | 1.98 4.29 5.59 | 1.00 3.16 4.07 | Hybrid | |
