As the prevalence of outsourcing maintenance activities by large mining houses increases, the cost model of these agreements, as well as the associated performance obligations, of the maintenance provider has become more complex, a situation which has been compounded in recent times with the sudden downturn in the global economy forcing all parties to become ever more efficient. This paper describes a study carried out to determine the possibility and efficacy of using simple, easy to understand, statistical methods to predict, to within a predefined range, the probable cost of operating a unit of trackless mining equipment as well as probable expected performance. It is concluded that there is definite benefit to be gained for the organisation in the ability to cost maintenance contracts based on risk rather than the current model of utilising actual, raw historical operating cost. The proposed method of a Wiebull analysis also shows promise as being a viable tool to with which to perform this function. Predicting availability however, due to the myriad of influencing factors, proves to be more difficult, although the structured approach proposed does provide a more effective solution than currently employed methodologies.