A New Model For Blast Fragmentation Prediction Based On Multivariate Analysis - Introduction

Several studies have been conducted on blastability and prediction of fragmentation. The term blastability is used to refer the susceptibility of the rock to blasting and is closely related to fragmentation. Parameters that determine fragmentation by blasting may be divided into two groups. The first group consists of blasting design parameters. The blast design parameters are controllable parameters and determined by the explosive engineers. The burden, spacing between holes, bench height, drillhole diameter, stemming height and explosive density are all design parameters. The second group consists of rock mass parameters that are referred to as uncontrollable parameters. The uncontrollable parameters are physical and mechanical properties of rock and rock mass structure. The strength, modulus of elasticity, density of rock, number of joint sets, joint orientation and joint spacing are among uncontrollable parameters. The rock mass parameters and blast design parameters should be considered together to explain fragmentation process. In this study, a blast database was accumulated to use for multivariate analysis. The blast database contains the blasts performed near Istanbul region quarries and the blasts that were preformed in various mines in the world by the blasting researchers. The cluster analysis was used to separate the blast data into different similarity groups. Group memberships were analyzed and checked by the discriminant analysis. Then, the multivariate regression analysis was used to develop a prediction equation for blast fragmentation for each group. Finally, a validation of the suggested models was performed for various mines. The model comprises blast design parameters, mechanical properties of rock and rock mass structure.