Mine safety management: An application of risk analysis, forecasting techniques, and Markov process to injury experience data

The purposes of this thesis research are the following: (i) identification of safety problems in a mine based on the analysis of accidents/injuries specific to the mine, (ii) evaluation of mine safety performance with such indices as risk of mine accidents and hazard rate, (iii) forecasting of accident rates, (iv) estimation of cost of accidents, and (v) cost-benefit analysis of proactive accident prevention programs. The quantitative procedures have been developed in this thesis for risk analysis, forecasting, and cost-benefit analysis. They have been applied to accident/injury experience data from three sample mines.; An analysis of the mine-specific accident/injury experience data using Safety and Health Technology Center (SHTC) database of Mine Safety and Health Administration reveals that some safety problems are generic and some are mine-specific. The application of a reliability technique to accident/injury experience data reveals that Weibull distribution is an appropriate model for describing the distribution of time-between-accidents. An investigation of the time series models and their application to mine-specific accident/injury data reveal that more complex Box-Jenkins's ARMA model as well as the first-order autoregressive model do not give better results than the simple exponential smoothing model. An application of the discrete-time, discrete-state space Markov model to accident/injury experiences reveals that it is reasonable to use this model to represent the stochastic behavior of a mine safety system. The U.S. Bureau of Mines' Accident Cost Indicator mode (ACIM) used in this thesis provides a common basis to calculate the cost of accidents and to perform a cost-benefit analysis of accident prevention programs. The ACIM costs, along with the Markov decision model, provide a suitable framework for the cost-benefit analyses.; This thesis significantly contributes to the objective of increasing the health and safety performance of the mining industry through the investigation and application of quantitative techniques to mine-specific accident/injury experience data. The quantitative procedures developed in this thesis will aid mine safety management personnel to develop focused accident prevention strategies including training, as well as to justify the investment in such proactive programs through cost-benefit analyses.