LAUSR

Abstract Buried concrete pipes are load bearing structures that need to resist the loads imposed by the surrounding ground. The common approach to design buried concrete pipes is based on an empirical method called the Indirect Design Method, which uses the laboratory capacity of the buried pipe linked to the field capacity using an empirical factor known as the bedding factor. However, limited published studies have investigated this bedding factor or tried to improve the current bedding factor values. Therefore, this study investigated the bending moment and bedding factor for concrete pipes under soil loads by conducting a parametric study investigating the effect of the installation condition, pipe diameter, pipe thickness and backfill height. A validated finite element model has been used for this purpose. The bedding factors obtained from the analyses have been compared with the bedding factors currently adopted by the AASHTO and British Standard (BS) design standards. The results showed that the BS design standard is conservative. In addition, the AASHTO design standard has been shown not to be safe for pipes with a diameter of 0.3 m and becomes more conservative as the diameter increases or the installation quality decreases. Therefore, new bedding factor models have been proposed using the results of the finite element modelling utilising an evolutionary polynomial regression (EPR) method. The paper demonstrates that the new models could be used for the economic and robust design of concrete pipes. The proposed models in this paper have the potential to significantly reduce the costs involved in either construction or maintenance of buried concrete pipes.