Optimal design of piled foundations using relative difference quotient algorithm

This paper presents the application of relative difference quotient algorithm (RDQA) to the least cost design of bored piled foundations. The objective function is the combined costs of soil excavation, pile cap, piles, and soil backfill. The design variables, including pile length, pile diameter, depth of pile cap, pile spacing, and pile number, are all discrete. RDQA is a local search method. It was developed based on the assumption that the objective and constraint functions are all monotonic functions. However, the problem of a piled foundation design belongs to a multi extreme values problem. The objective and constraint functions are not monotonic ones. Therefore, a modified RDQA (MRDQA) searching procedure and a strategy for determining the initial design are proposed in this paper. The efficiency and validity of MRDQA will be verified by comparing the solutions with the global optimum solutions obtained from exhaustive search method (ESM). The comparative results of two cases have shown that the errors of the solutions obtained by MRDQA are around 1.69% and 0.00%, respectively.