Author ORCID Identifier

Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Mining Engineering

First Advisor

Dr. Zach Agioutantis


Pillars form an important support structure in any underground mine. A bulk of the overburden load is borne by the mine pillars. Thus, the strength of pillars has been a subject of detailed research over more than 6 decades. This work has led to the development of largely empirical pillar design formulations that have reduced the risk of pillar failures and mine collapse. Current research, however, has drawn attention to the fact that some of the assumptions used in the development of conventional pillar design methodologies are not always valid. Conventional pillar design methodology assumes that the pillars carry the dead weight of the overburden. This conventional method treats the pillars as passive structures. The limitation of this approach is that the self-supporting capacity of the overburden is not incorporated in pillar design. This suspension theory of pillar design treats the strata-pillar interaction problem as a classic case of static equilibrium, without detailing the interactions of the two structures.

Globally, multiple pillar design methods have been developed, based on this suspension theory. Each of these methods approaches the calculation of pillar stability a little differently with respect to material properties, excavation geometries and stress conditions. Most of these design methods are derived empirically and lack a mechanics-based approach. Moreover, there is a lack of a unified pillar design methodology that can be used to design all types of mine pillars using a mechanics-based approach.

The Ground Reaction Curve has been used as a means of correlating strata displacements to stress conditions. In addition, the Support Reaction Curve has been used in modeling the response of a support system under load, as a function of support properties and installation time with respect to opening development. In comparing the Ground Reaction Curves and Support Reaction Curves for different support systems, one can evaluate the effectiveness of installed support systems in maintaining the integrity of the excavated area(s).

This approach has been widely used in designing secondary (artificial) support systems in both civil tunneling and the mining industry. Encouraged by the successful use of this single method in designing secondary support systems, this research revisits this concept for mine pillar design. This research investigates the utilization of the Ground Reaction Curve and Support Reaction Curve for the design of mine pillar support systems with respect to anticipated pillar loading and opening convergence. In addition, a conceptual three-tier solution to the pillar design problem, using a proper combination of numerical, analytical and data-driven methods is suggested, and a flowchart for the pillar design methodology is proposed. At the focus of this proposed method lies the Ground Reaction Curve (GRC) Concept. This research effort tries to verify the proposed pillar design flowchart using in-mine instrumentation and numerical modeling.

For the purpose of this research, a deep longwall coalmine is instrumented to measure changes in pillar stress and associated roof convergence, due to mining activity. Subsequently, numerical models were developed in FLAC3D to model the geomechanical effects of underground longwall mining. The numerical modeling results are validated and calibrated using instrumentation data and a surface subsidence profile. The calibrated numerical models are further used to generate the Ground Reaction Curve for the overburden and Support Reaction Curve for the coal pillar. The comparison of both curves gives a detailed view of the overburden stability with respect to the mine pillar loading, in a more mechanics-based sense. The developed numerical approach can be used in future research and further development of this methodology for various mine types and different pillar support systems.

Digital Object Identifier (DOI)

Funding Information

A significant part of this study was sponsored by the Alpha Foundation for the Improvement of Mine Safety and Health, Inc. (ALPHA FOUNDATION).

(Solicitation: AFC719; 2018-2020)