Table 4 Prediction model performance in a single project (BBDM project) database.

From: Machine learning approach for prediction of TBM performance and risk of jamming in Himalayan geology using a cross-project tunnelling database

Model

Training Set

Testing Set

R2

MAE

RMSE

MAPE

R2

MAE

RMSE

MAPE

XGBoost

0.957

2.018

4.174

3.039

0.936

2.187

3.996

3.410

RF

0.980

1.292

2.196

2.000

0.936

2.397

3.981

3.743

Bagging

0.971

1.501

2.651

2.303

0.936

2.331

3.968

3.634

Stacking

0.962

1.654

2.610

2.477

0.938

2.398

3.929

3.753

ANN

0.964

1.475

2.494

2.239

0.938

2.321

3.925

3.587

Back to article page