The solid-water fluidized bed was investigated with a binary mixture of irregularly shaped sand particles. A binary mixture was produced by mixing particles of sand for different weight ratios. The influence of various operating parameters on minimum elutriation velocity was investigated. It was observed that the Ume decreases with the increase of the lighter particles in the binary mixture, and the Ume increases with the increase of column diameter. A simplified empirical correlation has been developed to predict minimum elutriation velocity with acceptable statistical parameters. Application concerning a hybrid of artificial neural network (ANN), and genetic algorithm (GA), is successfully predicted.