Cost of foundation failures due to limited site investigations

Abstract

One of the greatest causes of foundation failure is due to insufficient knowledge of ground conditions. Uncertainty in ground conditions can also cause significant cost overruns and time delays for both client and contractor. Site investigations aim to reduce the uncertainty of ground conditions by various combinations of field and laboratory testing. However, time and cost constraints, as well as the judgement and experience of the consulting geotechnical engineer, have traditionally governed the scope of such site investigations. Analyses have been undertaken to investigate the performance of various site investigation schemes with respect to the cost of the resulting pad foundation system and the probability of failure. Penalty costs are attributed to foundation designs that experience excessive settlement to enable direct comparisons with foundation designs that conform to the design criteria. Designs resulting from the site investigation data are compared to a benchmark design achieved by employing a 3-dimensional finite element analysis within a trial-and-error process. Many realisations are undertaken for varying soil types in a Monte Carlo analysis by generating 3-dimensional simulated soil profiles using random field theory. The results illustrate a decreasing trend of total foundation cost for an increasing site investigation scope. The results also show that the cost of a foundation, excluding the penalty cost of failure, designed using an increased amount of knowledge regarding the site, does not always result in a less expensive foundation. However, all results suggest that a site investigation scheme with limited testing will result in a more expensive foundation, when the cost of possible foundation failure is included.