An Approach to Estimate the Probability in Meeting Army Maintenance Personnel Requirements

Estimating the maintenance workforce requirements of the Australian Army is a process of matching maintenance demand, resources, cost, and risk. Analysis of the equipment-specific demand has been historically undertaken within Army using a static mean-based estimate of tradesperson effort, known as work value units (WVU). An alternative to the mean-based approach is to consider a quantile methods where the analyst can either calculate the probability of meeting liabilities over the course of the year with a fixed capability, or conversely determine the number of hours to achieve a certain probability of meeting a fixed fleet liability. A difficulty of this approach is that the annual maintenance liability for different equipment types are represented by the sum of non identical random variables and an analytical description of a distribution of this type is impossible to obtain in general. This paper proposes a bootstrap simulation approach and an approximated analytical distribution to estimate the annual total maintenance time for a range of different equipment types. From this we can estimate the probability in not being able to support multiple fleets of arbitrary size, and hence allow a baseline for workforce development.