Rapid Shear Capacity Prediction of TRM-Strengthened Unreinforced Masonry Walls through Interpretable Machine Learning using a Web App

TL;DR

This study develops an interpretable machine learning tool integrated into a web app for rapid and accurate prediction of shear capacity in TRM-strengthened unreinforced masonry walls, aiding engineers and researchers.

Contribution

It introduces a data-driven, ensemble machine learning approach with interpretability and a user-friendly web app for shear capacity estimation of masonry walls.

Findings

Final model achieves R^2 of 0.95 and 8.03% MAPE.

Key predictors identified as $A_m$, $f_t$, and $n\cdot t_f$.

Web app facilitates easy access for practitioners.

Abstract

The presented study aims to provide an efficient and reliable tool for rapid estimation of the shear capacity of a TRM-strengthened masonry wall. For this purpose, a data-driven methodology based on a machine learning system is proposed using a dataset constituted of experimental results selected from the bibliography. The outlier points were detected using Cook's distance methodology and removed from the raw dataset, which consisted of 113 examples and 11 input variables. In the processed dataset, 17 Machine Learning methods were trained, optimized through hyperparameter tuning, and compared on the test set. The most effective models are combined into a voting model to leverage the predictive capacity of more than a single regressor. The final blended model shows remarkable predicting capacity with the determination factor ($R^2$) equal to 0.95 and the mean absolute percentage error equal to 8.03\%. In sequence, machine learning interpretation methods are applied to find how the predictors influence the target output. $A_m$, $f_t$, and $n\cdot t_f$ were identified as the most significant predictors with a mainly positive influence on the shear capacity. Finally, the built ML system is employed in a user-friendly web app for easy access and usage by professionals and researchers.