Accurate estimation of bearing capacity of stone columns reinforced: An investigation of different optimization algorithms

The precise estimation of bearing capacity (qrs) of stone columns reinforced with geogrid is crucial given the intricate nature of geotechnical materials and geological factors. However, the cost and complexity involved in determining qrs necessitate the use of a precise and consistent nonlinear equation suitable for diverse case studies. To address this, intelligent methods like nature-inspired optimization algorithms have emerged as effective solutions, enabling time and resource savings through accurate modeling. This research explores the utilization of two optimization algorithms, specifically Artificial Bee Colony (ABC) and Harmony Search (HS), for the estimation of qrs. Input parameters for modeling encompass the ratios of geogrid-reinforced layer diameter to footing diameter, GRSB and USB thickness to base diameter, unreinforced soft clay qrs, stone column length to diameter, and settlement to footing diameter. Finally, to assess the precision of the models, statistical indicators including Variance Account For (VAF), squared correlation coefficient (R2), mean absolute percentage error (MAPE), mean square error (MSE), and root mean square error (RMSE) were computed. According to the findings of this study, the accuracy achieved by employing smart methods using the ABC algorithm ranged from 0.981 to 0.989, with error rates ranging from 7.86 x 10-5 to 0.00883. Similarly, the accuracy of the HS algorithm was determined to be between 0.984 and 0.988, with error rates ranging from 3 x 10-5 to 0.00551. These results underscore the high accuracy of intelligent algorithms, offering a dependable means of determining qrs across various study areas while considering uncertainties.