JOURNAL OF ROCK MECHANICS

JOURNAL OF ROCK MECHANICS

Development of an Adaptive Neuro-Fuzzy Inference System (ANFIS) for Intelligent Prediction of Tunnel Settlement and Evaluation of Predictive Accuracy Based on Convergence Monitoring Data

Document Type : Original Article

Authors
Sasan Ghorbani 1
Farhad SamimiNamin 2
Mahdi Mahmoodjanloo 2
1 Department of Mining Engineering/, Hamedan University of Technology/ Hamedan/ Iran.
2 Department of Mining Engineering, University of Zanjan, Zanjan, Iran
10.22034/irsrm.2025.5008060.7011
Abstract
Tunnel excavation inevitably disturbs the in-situ stress state, leading to ground deformation and surface settlement that may affect adjacent structures and infrastructure. Conventional empirical and numerical approaches often struggle to capture the highly nonlinear and coupled interactions among geological conditions, excavation parameters, and structural responses. Soft computing techniques, especially hybrid neuro-fuzzy systems, offer an alternative data-driven strategy capable of modeling such complex relationships with improved adaptability and predictive performance. In this research, a database comprising 68 tunnel cases with similar excavation methods, geometrical characteristics, and ground conditions was compiled. The initial Fuzzy Inference System (FIS) structure was generated using subtractive clustering to automatically extract rule bases from the data. Gaussian membership functions were employed to ensure smooth transitions between fuzzy regions. The ANFIS model was trained in MATLAB using a hybrid learning algorithm that integrates least-squares estimation for consequent parameters and back-propagation for premise parameter tuning. The dataset was randomly divided into training, testing, and validation subsets to evaluate generalization capability. Model performance was assessed using statistical indices including mean error and correlation coefficient between predicted and measured settlements derived from convergence monitoring instruments. The developed ANFIS model demonstrated high predictive accuracy, with very low mean errors in both testing and validation phases and a strong correlation between predicted and observed settlement values. The results confirm the ability of the proposed framework to effectively model complex nonlinear relationships governing tunnel-induced deformation. The hybrid learning strategy contributed to stable convergence and minimized overfitting, while the rule-based fuzzy structure provided partial interpretability of the system behavior. Overall, the study indicates that ANFIS constitutes a reliable and efficient tool for intelligent settlement prediction and can support engineering decision-making in tunnel design, monitoring, and risk mitigation for projects with comparable geotechnical conditions.
Keywords
Neuro-Fuzzy Inference System
Tunnel settlement
Convergence meter
Network training
Subjects
[1] Zhu, Z., & Liu, C. (2025). A review of tunnelling caused ground surface settlement prediction with artificial intelligence method. Civil and Energy Research, 2-20. DOI: 10.55976/cer.1202513572-20
[2] Wei, G., Zhang, X. H., Xu, Y. F., & Wang, Z. (2019). Prediction of ground settlement due to excavation of a quasi-rectangular shield tunnel based on stochastic medium theory. Geotechnical and Geological Engineering, 37(5), 3605-3618. DOI: 10.1007/s10706-019-00856-7
[3] Zou, N., Cheng, H. Z., Dai, T., & Yang, Z. H. (2026). Real-time surface settlement prediction in shield tunneling using a time-series framework and interpretable machine learning. Results in Engineering, 109478. DOI: 10.1016/j.rineng.2026.109478
[4] Niu, G., & He, X. (2026). A Deep Learning–Based Approach for the Evaluation of Tunneling-Induced Ground Surface Settlement Considering Spatial Variability of the Soil. International Journal of Geomechanics, 26(3), 04026010. DOI: 10.1061/IJGNAI.GMENG-11938
[5] Xiao, Z., Zhou, X., Zhang, C., Xiao, Y., Wei, C., & Chen, J. (2025). Machine Learning-Based Prediction of Ground Settlement in Twin-Tunnel Excavation: A Case Study from Dongguan Metro Line 1. Indian Geotechnical Journal, 1-20. DOI: 10.1007/s40098-025-01456-4
[6] Gu, J., Zhou, J., & Li, C. (2026). Metaheuristic-optimized least squares boosting model for longitudinal settlement prediction in EPB shield construction. Tunnelling and Underground Space Technology, 169, 107326. DOI: 10.1016/j.tust.2025.107326
[7] Chen, S., Wang, Y., Huang, J., Xu, D., & Ravy, M. (2025). Dynamic Surface Deformation Prediction in Urban Shield Tunneling Based on Hybrid Machine Learning Method. Transportation Infrastructure Geotechnology, 12(7), 257. DOI: 10.1007/s40515-025-00718-1
[8] Li, S., Lan, A., Yin, L., Yu, B., & Sun, Y. (2026). Prediction of Tunnel Deformation Based on Machine Learning and Numerical Simulation. Indian Geotechnical Journal, 1-17. DOI: 10.1007/s40098-026-01479-5
[9] Yang, P., Zhou, J., Zhang, Y., Xu, C., Khandelwal, M., & Huang, S. (2025). Ground settlement prediction in urban tunnelling: leveraging metaheuristic-optimized random forest models. Arabian Journal for Science and Engineering, 50(20), 17025-17044. DOI: 10.1007/s13369-025-10062-8
[10] Ye, X. W., Jin, T., & Chen, Y. M. (2022). Machine learning-based forecasting of soil settlement induced by shield tunneling construction. Tunnelling and Underground Space Technology, 124, 104452. DOI: 10.1016/j.tust.2022.104452
[11] Tang, L., & Na, S. (2021). Comparison of machine learning methods for ground settlement prediction with different tunneling datasets. Journal of Rock Mechanics and Geotechnical Engineering, 13(6), 1274-1289. DOI: 10.1016/j.jrmge.2021.08.006
[12] Chen, R., Zhang, P., Wu, H., Wang, Z., & Zhong, Z. (2019). Prediction of shield tunneling-induced ground settlement using machine learning techniques. Frontiers of Structural and Civil Engineering, 13(6), 1363-1378. DOI: 10.1007/s11709-019-0561-3
[13] Wang, Q., Shen, C., Tang, C., Guo, Z., Wu, F., & Yang, W. (2024). Machine learning-based forecasting of ground surface settlement induced by metro shield tunneling construction. Scientific Reports, 14(1), 31795. DOI: 10.1038/s41598-024-82837-2
[14] Zhang, P., Wu, H. N., Chen, R. P., & Chan, T. H. (2020). Hybrid meta-heuristic and machine learning algorithms for tunneling-induced settlement prediction: A comparative study. Tunnelling and Underground Space Technology, 99, 103383. DOI: 10.1016/j.tust.2020.103383
[15] Liu, L., Zhou, W., & Gutierrez, M. (2023). Physics-informed ensemble machine learning framework for improved prediction of tunneling-induced short-and long-term ground settlement. Sustainability, 15(14), 11074. DOI: 10.3390/su151411074
[16] Chen, L., Hashiba, K., Liu, Z., Lin, F., & Mao, W. (2023). Spatial-temporal fusion network for maximum ground surface settlement prediction during tunnel excavation. Automation in Construction, 147, 104732. DOI: 10.1016/j.autcon.2022.104732
[17] Naeini, S. A., & Khalili, A. (2017). Settlement Prediction for Tehran Subway Line-3 via FLAC3D and ANFIS. International Journal of Geotechnical and Geological Engineering, 11(7), 612-617.
[18] Jang, J. S., & Sun, C. T. (1995). Neuro-fuzzy modeling and control. Proceedings of the IEEE, 83(3), 378-406. DOI: 10.1109/5.364486
[19] Liu, X., Hussein, S. H., Ghazali, K. H., Tung, T. M., & Yaseen, Z. M. (2021). Optimized Adaptive Neuro‐Fuzzy Inference System Using Metaheuristic Algorithms: Application of Shield Tunnelling Ground Surface Settlement Prediction. Complexity, 2021(1), 6666699. DOI: 10.1155/2021/6666699
[20] Vieira, J., Dias, F. M., & Mota, A. (2004). Neuro-fuzzy systems: a survey. In 5th WSEAS NNA international conference on neural networks and applications, Udine, Italia, 1-6.
[21] Jang, J. S. (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE transactions on systems, man, and cybernetics, 23(3), 665-685.
[22] Nikolić, V., Petković, D., Shamshirband, S., & Ćojbašić, Ž. (2015). Adaptive neuro-fuzzy estimation of diffuser effects on wind turbine performance. Energy, 89, 324-333. DOI: 10.1016/j.energy.2015.05.126
[23] Kourehli, S. S., Chehre, M. K., & Zamani, A. G. (2016). Prediction of structural damage location with adaptive neuro-fuzzy inferential system. Iranian Journal of Structural Engineering, 3(1), 61-71
JOURNAL OF ROCK MECHANICS
Volume 9, Issue 3
Autumn 2025
Pages 61-79