It has long been postulated that a relationship exists between commodity price cycles and fatalities in the mining industry.Previous studies have found only weak correlations in this area.This study analyses the fatal...It has long been postulated that a relationship exists between commodity price cycles and fatalities in the mining industry.Previous studies have found only weak correlations in this area.This study analyses the fatalities recorded in coal mines over the period 1985-2016 in the State of Queensland as a function of thermal coal price variation.The study finds that the relationship between fatalities and coal prices is not linear.One to two fatalities occur in most years independent of the thermal coal price.When the price of coal falls below AUD 55/tonne(non-inflation adjusted),the likelihood of an incident involving multiple fatalities increases.The probability can be estimated at 2 in 18 events(equivalent to 11%).This paper postulates that in difficult economic times,mining companies react by downsizing direct employees.If not carefully managed,this can result in loss of knowledge around safety systems,and reduced effectiveness of safety supervision.Because of labour cost advantages,some jobs previously undertaken by direct employees will be replaced by contractors.Increased contractor numbers contribute to increased risk of fatalities occurring,as contractors are over-represented in accident categories involving vehicle accidents,tire handling and crushing incidents.Mine inspectorates,mining,and mining contractor companies need to be especially vigilant to enforce health and safety management systems during periods of low coal prices.展开更多
To solve information asymmetry problem on online auction, this study suggests and validates a forecasting model of winning bid prices. Especially, it explores the usability of data mining approaches, such as neural ne...To solve information asymmetry problem on online auction, this study suggests and validates a forecasting model of winning bid prices. Especially, it explores the usability of data mining approaches, such as neural network and Bayesian network in building a forecasting model. This research empirically shows that, in forecasting winning bid prices on online auction, data mining techniques have shown better performance than traditional statistical analysis, such as logistic regression and multivariate regression.展开更多
文摘It has long been postulated that a relationship exists between commodity price cycles and fatalities in the mining industry.Previous studies have found only weak correlations in this area.This study analyses the fatalities recorded in coal mines over the period 1985-2016 in the State of Queensland as a function of thermal coal price variation.The study finds that the relationship between fatalities and coal prices is not linear.One to two fatalities occur in most years independent of the thermal coal price.When the price of coal falls below AUD 55/tonne(non-inflation adjusted),the likelihood of an incident involving multiple fatalities increases.The probability can be estimated at 2 in 18 events(equivalent to 11%).This paper postulates that in difficult economic times,mining companies react by downsizing direct employees.If not carefully managed,this can result in loss of knowledge around safety systems,and reduced effectiveness of safety supervision.Because of labour cost advantages,some jobs previously undertaken by direct employees will be replaced by contractors.Increased contractor numbers contribute to increased risk of fatalities occurring,as contractors are over-represented in accident categories involving vehicle accidents,tire handling and crushing incidents.Mine inspectorates,mining,and mining contractor companies need to be especially vigilant to enforce health and safety management systems during periods of low coal prices.
文摘To solve information asymmetry problem on online auction, this study suggests and validates a forecasting model of winning bid prices. Especially, it explores the usability of data mining approaches, such as neural network and Bayesian network in building a forecasting model. This research empirically shows that, in forecasting winning bid prices on online auction, data mining techniques have shown better performance than traditional statistical analysis, such as logistic regression and multivariate regression.
