Abstract:
Gene Expression Programming is employed to predict the 28 days compressive strength of cement mortar. The input parameters considered are C3S, SO3, Blaine and Alkali and the output parameter is the 28 days compressive strength. The model was able to predict successfully with a root mean square error of 1.4956. This model is compared with the Fuzzy Logic Model and ANN-GA model. The GEP model is proved to perform better than the Fuzzy Logic Model. It yields an expression that relates the inputs to outputs thereby overcoming the disadvantages of the artificial neural networks.