Optimized Adaptive Neuro-Fuzzy Inference System Using Metaheuristic Algorithms: Application of Shield Tunnelling Ground Surface Settlement Prediction

Deformation of ground during tunnelling projects is one the complex issues that required to be monitored carefully avoid unexpected damages and human losses. Accurate prediction settlement (GS) a crucial concern for problems, adequate predictive model can vital tool tunnel designers simulate accurately. This study proposes relatively new hybrid artificial intelligence (AI) models predict earth pressure balance (EPB) shield in Bangkok MRTA project. The were various nature-inspired frameworks, such as differential evolution (DE), particle swarm optimization (PSO), genetic algorithm (GA), ant colony optimizer (ACO) tune adaptive neuro-fuzzy inference system (ANFIS). To obtain accurate reliable results, modeling procedure established based on four different dataset scenarios including (i) preprocessed normalized (PPN), (ii) nonnormalized (PPNN), (iii) non-preprocessed (NPN), (iv) (NPNN) datasets. Results indicated PPN scenario significantly affected terms their perdition accuracy. Among all developed models, ANOFS-PSO achieved best predictability performance. In quantitative terms, PPN-ANFIS-PSO attained least root mean square error value (RMSE) 7.98 correlation coefficient (CC) 0.83. Overall, results confirmed superiority explored AI robust tunnelling.