Journals Information
Civil Engineering and Architecture Vol. 13(5), pp. 4019 - 4030
DOI: 10.13189/cea.2025.130540
Reprint (PDF) (2321Kb)
The Optimization of Predicting the Bearing Capacity of Bored Piles Using the Finite Element Method and the Artificial Neural Network
Truong Xuan Dang 1, Phuong Tuan Nguyen 2, Tuan Anh Nguyen 3,*, Hoa Van Vu Tran 3
1 Department of Urban Infrastructure Management, Ho Chi Minh City University of Natural Resources and Environment, Vietnam
2 Department of Construction, Mien Tay Construction University, Vinh Long Province, Vietnam
3 The SDCT Research Group, Ho Chi Minh City University of Transport, Vietnam
ABSTRACT
This study aims to develop a model combining the Finite Element Method and Artificial Neural Network to predict the ultimate bearing capacity of bored piles under various geological conditions. FEM is employed to simulate the detailed interaction between piles and the soil foundation under applied loads, while ANN is trained using FEM-generated data to optimize prediction accuracy. This integrated model not only delivers precise predictions of bearing capacity but also constructs a comprehensive bearing capacity distribution map for the entire study area, significantly aiding in pile foundation design. The findings demonstrate that the FEM-ANN model outperforms traditional methods and standalone models in terms of accuracy. The Mean Absolute Error (MAE) metrics for ANN highlight reliable prediction capabilities, particularly under heavy load conditions. Comparative analysis of results from FEM, ANN, and experimental data confirms that this integration reduces prediction errors and leverages the strengths of both approaches. The ultimate bearing capacity map developed by the model effectively captures the load distribution across the study area, enabling engineers to identify optimal pile locations and reduce construction costs. In conclusion, the integration of FEM and ANN not only achieves high accuracy but also proves to be practically effective, particularly in complex infrastructure projects.
KEYWORDS
Finite Element Method (FEM), Artificial Neural Networks (ANN), Bored Pile Bearing Capacity, Predictive Modeling in Geotechnics, Infrastructure Foundation Optimization
Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Truong Xuan Dang , Phuong Tuan Nguyen , Tuan Anh Nguyen , Hoa Van Vu Tran , "The Optimization of Predicting the Bearing Capacity of Bored Piles Using the Finite Element Method and the Artificial Neural Network," Civil Engineering and Architecture, Vol. 13, No. 5, pp. 4019 - 4030, 2025. DOI: 10.13189/cea.2025.130540.
(b). APA Format:
Truong Xuan Dang , Phuong Tuan Nguyen , Tuan Anh Nguyen , Hoa Van Vu Tran (2025). The Optimization of Predicting the Bearing Capacity of Bored Piles Using the Finite Element Method and the Artificial Neural Network. Civil Engineering and Architecture, 13(5), 4019 - 4030. DOI: 10.13189/cea.2025.130540.