Abstract:
This study uses machine learning to predict the compressive strength of cement-sand mortar
incorporating supplementary cementitious materials (SCMs). The research addresses a gap in the
literature by specifically examining how the oxide composition of SCMs influences mortar
strength. Using a dataset of various mortar mixes, several machine learning models were tested,
with the extreme gradient boosting (XGB) model emerging as the most effective, achieving a
testing R2 of 0.90. The results show that the curing period is the most influential factor on
compressive strength, followed by the oxide compositions of the SCMs. This work highlights the
potential of machine learning for enhancing material performance predictions and supports the
development of more sustainable and durable construction practices.
