Prediction of compressive strength of masonry structures: Integrating three optimized models by virtue of committee machine

An accurate computation of the compressive strength of masonry structures is an overarching factor in design and construction of masonry structures. This considerable significance compels researchers to propose an appropriate, reliable and, more generalized method whereby the precise value of compressive strength of ma-sonry structures is calculated. In the current study, a committee machine with optimized elements is constructed, thereby extracting a non-linear relationship between compressive strength of masonry structures with compressive strength of mortar and brick. In order to accomplish this objective, three intelligent models viz. neural network, fuzzy inference system, and support vector regression are firstly optimized with bat-inspired algorithm, and these improved models are subsequently applied for estimation of compressive strength of ma-sonry structures. bat-inspired algorithm is hybridized with intelligent models for extracting the best values of weights and biases of neural network, membership's functions of fuzzy inference system, and user-defined pa-rameters of support vector regression. Then, committee machine is utilized for amalgamating the outputs of three optimized models incl. optimized neural network, optimized fuzzy inference system, and optimized support vector regression. bat-inspired algorithm is also embedded in the structure of committee machine, thereby determining the optimal contribution of each optimized model in the final prediction. Data sets including 96 records of accessible in the literature are used to learn and evaluate the constructed models. Appraisal of the accuracy based on statistical parameters verified that the committee machine could effectively improve the prediction accuracy of the optimized models and also has a better performance compared to commonly well-known predictive correlations. This study also proved that committee machine with optimized elements is a very convenient approach for mapping nonlinear functions between compressive strength of masonry structures and compressive strength of brick and mortar.

Automated summary

Accurately calculating the compressive strength of masonry structures is crucial for design and construction. This study proposes a committee machine with optimized elements for estimating the compressive strength by extracting nonlinear relationships between the strength of masonry structures, mortar, and brick. The results show that the committee machine improves the prediction accuracy and outperforms commonly used correlations.