The average block size of rock after bench blasting in openpit mine is an important measurement for blasting quality.It is also significant for subsequent shoveling and transportation.In order to predict the average block size after bench blasting,the parameters of Support Vector Machine(SVM)were optimized using Fruit Fly Optimization Algorithm(FOA).By establishing FOA-SVM,the average block size of rock after blasting is predicted.The impact of traditional SVMs parameters selection on the accuracy of the prediction results is avoided.The results show that the FOA-SVM can achieve a better prediction of the average block size of rock blasting.