Using neural networks to predict workability of concrete incorporating metakaolin and fly ash

This paper details the development of neural network models that provide effective predictive capability in respect of the workability of concrete incorporating metakaolin (MK) and fly ash (FA). The predictions produced reflect the effect of graduated variations in pozzolanic replacement in Portland cement (PC) of up to 15% MK and 40% FA. The results show that the models are reliable and accurate and illustrate how neural networks can be used to beneficially predict the workability parameters of slump, compacting factor and Vebe time across a wide range of PC–FA–MK compositions.