Forecasting of Militrary Spare Part Demands

Jasmine Chee Shi Ying

Abstract

Large-scale fleet management organisation, like the military, often stores equipment and spare parts at many different sites. Due to environmental and field operating conditions differences, equipment deployed at different operating sites has different observed failure rates and hence spare part demands. In addition, as the equipment is introduced into operation in phases, the equipment tends to exhibit different reliability. Furthermore, with different usage intensity, it will result in a spread of equipment ages in the long run.

While the case of homogenous equipment failure characteristic have been popular in the modelling of spare part demands, it is not appropriate in many military systems, where deployment of the fleet is dispersed over a multiplicity of operating environment. For these scenarios, models involving mixtures of failure distribution becomes appropriate. In this thesis, we proposed a demand forecasting system for spare-parts management in large-scale fleet scenarios with substantial variability in equipment reliability. In fact, these models are not only applicable to the military, but also to other industries such as transportation companies, and spare parts maintenance which have similar business characteristics.