Multi-algorithm optimization of palm oil fuel ash dosage and curing duration for Maximum dry density prediction in sustainable soil stabilization

This study evaluates ten optimization algorithms to determine the optimum palm oil fuel ash (POFA) dosage and curing time for soil stabilization. Differential Equation Optimization performed best, identifying an optimum POFA dosage of 5% and a curing period of 7 days. For maximum dry density prediction, it achieved a mean absolute percentage error (MAPE) of 1.02%, a root mean square error (RMSE) of 0.011, and an R2 of 0.991. Simulated Annealing and Fibonacci Sequence Optimization also demonstrated competitive performance, with MAPEs of 1.46% and 1.28%, respectively. Collatz-Inspired Search produced the highest MAPE of 6.01%. A strong positive correlation (r = 0.92) between excess POFA dosage and prediction error emphasizes the importance of dosage selection for effective soil stabilization.