Predicting Cement Compressive Strength Using Double-Layer Multi-expression Programming

This paper presents a novel algorithm named Double-layer Multi-expression Programming (DMEP). Then DMEP model is applied to the prediction of 28-day Portland cement compressive strength. We compare DMEP model with other four soft computing models, namely Multi-Expression Programming model (MEP), Gene Expression Programming model (GEP), Neural Networks model (NN) and Fuzzy logic model (FL) on 28-day cement strength prediction. The experiment results show that DMEP model has a lower rate in RMSE and MAE, and the prediction average percentage error of the cement strength data using DMEP model is reduced to 2.18%, which is lower than other four models. The results obtained from the computational tests demonstrate that DMEP is a promising technique for the prediction of cement strength.