Machine learning based predictive insights into compressive strength and strength retention of magnesium oxychloride cement

Magnesium oxychloride (MOC) cement is a sustainable alternative to Portland cement (PC) due to its excellent mechanical properties, but the poor water resistance limits its usage. Various supplementary cementitious materials (SCMs) and chemical additives can enhance MOC's strength retention (SR) after exposure to water. However, MOC cement design to attain desirable compressive strength (CS) and SR requires a precise selection of input materials, mix proportions, and additives, which is quite challenging. Therefore, employing machine learning techniques could facilitate and accelerate the development of MOC binders with the desired properties. So, this study employed four machine learning (ML) models to predict the CS and SR of MOC using two literature-derived databases. This study evaluates the performance of various ML models for accurate CS and SR prediction of MOC cement. The GBR and XGB models emerged as the most effective predictors for both CS and SR datasets, achieving the best prediction accuracy (coefficient of determination; R 2 ≥0.90) on testing data. In addition, Shapely Additive Explanation (SHAP) is used to interpret the effect of various parameters on CS and SR and is discussed in detail. SHAP-derived feature contributions and experimental observations from existing literature are consistent. The interpretable ML strategy used in this study will aid in the production and performance tuning of CS and SR of sustainable MOC cement for widespread applications.