Mark John Masia | University of Newcastle - Austrália

Abstract

The Australian masonry design code (AS3700-2001) has been in a limit states format since 1988. However, limit state specifications for masonry design in Australia, US, Canada and Europe have not been developed from reliability-based calibration methods but rather calibrated to past practice. The Brazilian Code NBR10837 is nowadays under revision. The main change will be the adoption of the Ultimate Limite States for the guarantee of safety. Although it is commonly believed that current design models are conservative, the actual level of safety of masonry structures is not known. It is unknown how structures designed to the masonry design codes compare to structures designed using other materials in terms of reliability (or safety) and also if different masonry walls and other structural elements have similar levels of reliability. The problem is compounded by the fact that the strength properties of masonry are highly variable. Masonry is a complex material consisting of brick/block units set in a more flexible mortar matrix. The mortar joints act as planes of weakness due to the inherently low bond strength between units and mortar. High unit-to-unit spatial variability is also observed, particularly for flexural tensile and torsional shear bond strengths, due to variations in the quality of workmanship, the weather during construction, and materials from location to location, all within the one structure. Very few studies have considered computational methods for calculating the structural reliability of masonry structures and these have been limited to the simplest loading condition only; namely, compression loading (Turkstra, 1989; Ellingwood, 1981). However, Stewart and Lawrence (2002) and Drake et al. (2004) have developed preliminary ‘proof-of-concept’ techniques to estimate the structural reliability of masonry walls for vertical one-way bending (i.e., transverse loading) and compression loading. Stewart and Lawrence (2002) were able to show the important effect unit-to-unit spatial variability can have on strength prediction and structural reliability. A need clearly exists for developing theoretical and computational models to enable the accurate and efficient calculation of reliability for new and existing masonry structures. Although extensive work has been directed at developing predictive strength models for masonry, very little effort has so far been directed towards the issues of model error (degree of accuracy of the predictive strength models), material behavior (unit strength) uncertainty and variability, unit-to-unit spatial strength variability and their effect on wall strength and reliability. A reliability analysis requires probabilistic information for all of these variables. A research project has commenced at The University of Newcastle, Australia. This project is jointed funded by the Australian Research Council and the clay brick industry in Australia. The main objective of the project is to develop computational and stochastic models needed for the calculation of strengths and structural reliabilities for masonry walls in flexure and compression. (AU)