Estimating the continuous risk of accidents occurring in the South African mining industry

[摘要] ENGLISH ABSTRACT: Statistics from mining accidents expose that the potential for injury ordeath to employees from occupational accidents is relatively high. This studyattempts to contribute to the on-going efforts to improve occupational safetyin the mining industry by creating a model capable of predicting the continuousrisk of occupational accidents occurring. Model inputs include the timeof day, time into shift, temperatures, humidity, rainfall and production rate.The approach includes using an Artificial Neural Network (ANN) to identifypatterns between the input attributes and to predict the continuous risk ofaccidents occurring. As a predecessor to the development of the model, acomprehensive literature study was conducted. The objectives of the studywere to understand occupational safety, explore various forecasting techniquesand identify contributing factors that influence the occurrence of accidents andin so doing recognise any gaps in the current knowledge. Another objectivewas to quantify the contributing factors identified, as well as detect the sensitivityamongst these factors and in so doing deliver a groundwork for thepresent model.After the literature was studied, the model design and construction wasperformed as well as the model training and validation. The training andvalidation took the form of a case study with data from a platinum minenear Rustenburg in South Africa. The data was split into three sections,namely, underground, engineering and other. Then the model was trained andvalidated separately for the three sections on a yearly basis. This resultedin meaningful correlation between the predicted continuous risk and actualaccidents as well as the majority of the actual accidents only occurring whilethe continuous risk was estimated to be above 80%. However, the underground section has so many accidents, that the risk is permanently very high. Yet, theengineering and other sections produced results useful for managerial decisions.

[发布机构] Stellenbosch University