Rational Decision Framework for Designing Pile-Load Test Programs

Abstract

Abstract There is currently an inconsistency in the recommendations that are available in pile-design codes and practices regarding the required number of proof-load tests and the level of the proof loads for piles. In this paper, a pre-posterior decision-making framework is proposed to allow for selecting the optimal pile-load test program that would result in the maximum expected benefit to a project, while maintaining a target level of reliability in the pile design at the site. This proposed methodology is original, practical, and is based on site-specific information that is unique to any given project. The proposed methodology is based on a robust Bayesian approach that allows for updating the capacity distribution of piles at a site, given the results of the proof-load test program. The efficiency of the proposed decision framework is demonstrated by applying it on a practical design example that involves piles that are driven in a site consisting of medium-dense sand. Results indicate that: (1) the optimum proof-load level that results in the maximum benefit to the example project is 1.5 times the design load, (2) the optimum number of tests is a function of the number of piles (superstructure load) and the costs of the pile construction and testing, (3) as the number of piles in the site increases, the optimal required number of proof-load tests also increase, with the optimum number of pile-load tests being around 1 % to 2 % of the total number of piles at the site, and (4) the optimal number of pile-load tests increases as the cost of pile construction and installation increases and as the cost of implementing the pile test program decreases.