EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Artificial Intelligence Applied to Soil Compaction Control for the Light Dynamic Penetrometer Method

Jorge Rojas-Vivanco, José García, Gabriel Villavicencio, Miguel Benz, Antonio Herrera, Pierre Breul, German Varas, Paola Moraga, Jose Gornall, Hernan Pinto

Mathematics · 22.10.2025 · 13 (21) · 3359

DOI: 10.3390/math13213359 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mathematics, online erschienen am 22.10.2025. Verfasst von Jorge Rojas-Vivanco und 9 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: FONDEF ANID.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Compaction quality control in earthworks and pavements still relies mainly on density-based acceptance referenced to laboratory Proctor tests, which are costly, time-consuming, and spatially sparse. Lightweight dynamic cone penetrometer (LDCP) provides rapid indices, such as qd0 and qd1, yet acceptance thresholds commonly depend on ad hoc, site-specific calibrations. This study develops and validates a supervised machine learning framework that estimates qd0, qd1, and Zc directly from readily available soil descriptors (gradation, plasticity/activity, moisture/state variables, and GTR class) using a multi-campaign dataset of n=360 observations. While the framework does not remove the need for the standard soil characterization performed during design (e.g., W, γd,field, and RCSPC), it reduces reliance on additional LDCP calibration campaigns to obtain device-specific reference curves. Models compared under a unified pipeline include regularized linear baselines, support vector regression, Random Forest, XGBoost, and a compact multilayer perceptron (MLP). The evaluation used a fixed 80/20 train–test split with 5-fold cross-validation on the training set and multiple error metrics (R2, RMSE, MAE, and MAPE). Interpretability combined SHAP with permutation importance, 1D partial dependence (PDP), and accumulated local effects (ALE); calibration diagnostics and split-conformal prediction intervals connected the predictions to QA/QC decisions. A naïve GTR-average baseline was added for reference. Computation was lightweight. On the test set, the MLP attained the best accuracy for qd1 (R2=0.794, RMSE =5.866), with XGBoost close behind (R2=0.773, RMSE =6.155). Paired bootstrap contrasts with Holm correction indicated that the MLP–XGBoost difference was not statistically significant. Explanations consistently highlighted density- and moisture-related variables (γd,field, RCSPC, and W) as dominant, with gradation/plasticity contributing second-order adjustments; these attributions are model-based and associational rather than causal. The results support interpretable, computationally efficient surrogates of LDCP indices that can complement density-based acceptance and enable risk-aware QA/QC via conformal prediction intervals.

Förderung

  • FONDEF ANID

Zitieren

APA

Rojas-Vivanco, J., García, J., Villavicencio, G., Benz, M., Herrera, A., Breul, P., Varas, G., Moraga, P., Gornall, J., & Pinto, H. (2025). Artificial Intelligence Applied to Soil Compaction Control for the Light Dynamic Penetrometer Method. Mathematics, 13(21), 3359. https://doi.org/10.3390/math13213359

Vancouver

Rojas-Vivanco J, García J, Villavicencio G, Benz M, Herrera A, Breul P, et al. Artificial Intelligence Applied to Soil Compaction Control for the Light Dynamic Penetrometer Method. Mathematics. 2025;13(21):3359. doi:10.3390/math13213359

BibTeX
@article{RojasVivanco2025,
  author = {Rojas-Vivanco, Jorge and García, José and Villavicencio, Gabriel and Benz, Miguel and Herrera, Antonio and Breul, Pierre and Varas, German and Moraga, Paola and Gornall, Jose and Pinto, Hernan},
  title = {Artificial Intelligence Applied to Soil Compaction Control for the Light Dynamic Penetrometer Method},
  journal = {Mathematics},
  year = {2025},
  volume = {13},
  number = {21},
  pages = {3359},
  doi = {10.3390/math13213359}
}
Kostenlos starten →