Probabilistic Models for Mechanical Properties of Clay and Fly Ash Brick

Abstract

Material Strength enormously dictates the design philosophy of a structure as nonuniformity in strength is an inherent property due to various known and unknown factors. Although, in case of concrete and steel, uncertainty aspect in strength is well researched but the same is absent for bricks. In the present study, around 500 clay and fly ash bricks considered and random variables in the form of shear bond strength of prepared specimens and compressive strength of individual units are measured and analyzed on a larger statistical domain for a large number of samples in terms of their variability properties, an underlying feature of which is the distribution function. With the attained experimental test strength data pool, around sixty distributions considered for each strength parameter corresponding to a material type and mortar grade used, the most familiar best probabilistic distribution reflecting the fashion of a particular strength property of brick is proposed with the aid of certain Goodness of Fit (GOF) tests.
The present study is also an exemplification of the underlying principles and analytical observations in deciding the best fitting probabilistic distribution function pertaining to a variable strength parameter. The results obtained indicates the various proposed probabilistic distributions that best describe the characteristics of the available non-uniform experimental data for shear bond strength with three different grades of mortar along with the compressive strength of clay and fly-ash bricks and mortar cubes. With the aid of GOF tests followed by Survival analysis, Rayleigh and Generalised Extreme distributions are found to best describe the variability in shear bond strength for clay and fly ash bricks while Log-Logistic and Generalised Extreme distributions are found to describe best the variability feature in case of compressive strength for clay and fly ash brick units. Beta, Normal and Weibull have been seemed to best describe the variability in compressive strength for the prepared mortar specimens. Hence, Beta, VI Normal and Weibull are proposed for describing the variability in compressive strength for CM1, CM2 and CM3 types of mortar grades.
The characteristic strengths are also obtained for all strength parameters from the survival analysis of the proposed probabilistic models.