Prediction of ground vibration is of great importance in mitigation of blast-induced adverse effects. In this research, an innovative multivariate analysis procedure for prediction of blast-induced ground vibration is presented. Several blasts performed in a sandstone quarry were brought together and an extensive blast database is created. Blast design parameters, and the parameters related to ground vibration were considered to perform multivariate analysis. The hierarchical cluster analysis was used to separate the blasts data into different groups of similarity. Group memberships were analyzed and confirmed by the discriminant analysis. A linear discriminant equation was created to classify prospective blasts. Regression analysis was applied to develop prediction equations for the estimation of blast induced ground vibration. Validation analysis of the proposed equations was performed using test blasts. Prediction capability of the proposed models was found to be strong. (C) 2012 Elsevier Ltd. All rights reserved.