Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Mining Engineering

First Advisor

Dr. Steven Schafrik


Movement of overburden or ore from one location to another may be considered the primary physical process of mining. The means by which it is transported is then of primary concern both to economic production and cost control. The geologic characteristics of some mineral deposits, especially coal and phosphates, lend themselves to overburden transport by means of cast blasting, which when efficiently designed has proven itself to be a cheaper method than traditional truck/shovel, dragline, or even dozer push transport. Simultaneously, this movement of a significant portion of overburden using explosives may enable an operation to increase mineral production, as existing equipment can be expected to retain its full productive capacity. Traditional cast blast design methodologies have relied largely on field observations from previous blasts, which may then be modified to current conditions. In recent decades, numerical methods have been applied to various aspects of the blasting process, including fracturing, movement, vibration, etc. The Discrete Element Modeling technique has been applied to the physical process of blasted rock movement. This method allows for modeling a large number of discrete particles, which may be effectively related to the rock fragments generated during a blast. Although significant progress has been made with this technique, further improvements may be made as computer processing capabilities progress, and new technologies and numerical modeling techniques become available. This dissertation seeks to explore the potential for several improvements in the assessment of the muck pile profile from cast blasting through the following:

a) The majority of discrete element modeling has used a spherical shape to describe rock fragments, for simplification of calculation. Some numerical modeling work has been done using uniform cubical (polyhedral) shapes. However, scale tests comparing modeling results with actual cast blast profiles are scarce. This research will explore the possibility of applying shape variability to the polyhedral elements used to populate the numerical model and compare the results with the measurement of actual cast blast events. Further, despite some researchers’ use of varying sizes of elements in their models, there is no clear correlation between size variation of the elements with actual blast fragmentation distributions. Also, no correlation of element size placement in the model to actual blast placement is evident. Modern drone and photogrammetric image analysis techniques will be applied to actual cast blasts to determine fragment distributions, which may then be used to calibrate a discrete element model designed to predict fragment movement.

b) A shortage of research publications in this field of DEM blast movement models incorporate a probabilistic approach to result prediction. It is proposed to apply a Monte Carlo scheme to various model parameters to create confidence intervals related to predicted results for muck pile profiles.

Digital Object Identifier (DOI)