An Evolutionary Polynomial Computing of Pile Capacity Using the Results of High-strain Dynamic Test

It is normal to do a pile load test to verify the pile capacity before construction using either a static load test or a high-strain dynamic test (HSDT). The latter is favourable due to its simplicity and lower cost, but it lacks a simple equation that could be used to interpret its results. This study aims to develop a simple model that could be easily utilized to estimate the pile capacity. The model has been developed using EPR-MOGA, a regression analysis aid with a genetic algorithm, utilizing database of HSDT of piles. The accuracy of the new model has been examined by calculating the mean absolute error, root mean square error, mean, coefficient of determination, and ratio of prediction with an error range of ± 20%. The model showed very good accuracy and thus, it could be used with confidence. The model has the potential to reduce costs and difficulties associated with the interpretation of the HSDT.