Genetic algorithm - Based optimum design of piled raft foundations with model tests

Abstract

In this study, a new approach using the genetic-based evolution algorithm without gradient requirements is proposed for the optimum design of piled raft foundations. The objective functions considered are differential settlements based on a serviceability of the structure and the cost-based total weight of the raft and piles. In the process of formulating the genetic algorithm-based optimum design procedure, the analysis of piled raft foundations is performed based on the 'hybrid' approach, considering interaction effects between pile-soil-pile, raft-soil-raft, and raft-soil-pile. The simple genetic algorithm is also used for the optimum design of piled raft foundations. Small-scaled laboratory model tests are further carried out to investigate the load-sharing ratios between piles and the raft. Also to verify the validity of the proposed genetic algorithm-based optimum design procedure, the optimal location of piles is compared with the result of small-scaled model tests.