This thesis addresses the topic of underground excavation stability in open stope mining methods. There are three fundamental aspects to be considered in an engineering rock mechanics design of open stopes. The first aspect is the characterization of the rock mass to identify and quantify the properties and components affecting the rock mass behaviour. The second aspect is the effect of the stress fields on the rock mass that may result in zones of high compressive stress or zones of relaxation in stope walls. The third aspect is the physical condition of the problem and is defined primarily by the size, geometry and relative orientation of openings with regard to the rock mass and the stress field. The interaction of these three fundamental aspects constitutes the problem to be investigated.

The principal objective of the study is to develop a reliable geomechanical model (based on the above three aspects) that can predict the stability of open stopes in typical Canadian geological settings. An empirical approach was chosen for the development of the model, because of the complexity of the problem and the difficulty in estimating the input parameters with precision. Empirical methods are likely to be more reliable since they make use of past experience. A considerable amount of effort has been spent in building a broad data base which includes more than 250 case histories of unsupported and supported stopes from 34 Canadian mines.

The application of the model in the back-analysis of a large number of representative case histories allowed calibration of each of the factors composing the model. Since the model's prediction corresponds very well to the actual stope behaviour in most case histories, the model is considered empirically verified.

The effect of external factors (parameters affecting stope stability that are not related to the geotechnical or geometrical conditions) have also been investigated. The limit of applicability of cable bolting has been defined and rough guidelines for the design of cable support systems are proposed based on the systematic compilation of past experience. The effect of blasting, although not quantified, has been observed in 18 case histories. More research is required in order to include the effect of blasting in the actual modelwhile the effect of time is not of concern in open stope mining, when there are no mining activities in the imediate area of the investigated stope.