Logistics support is a key element of aircraft transportation systems. This
paper is concerned with the impact of aircraft spares provisioning decisio
ns on the availability of aircraft. Spares provisioning in this context is
complicated by the fact that spares may be shared across aircraft and that
aircraft may have redundant systems. In addition, decisions concerning airc
raft spares support require a rapid response for safety reasons. Analytical
models have proven to provide a quicker response time than corresponding s
imulation models. There is an existing analytical model that includes the e
ffect of redundancy and spares, but the underlying assumption is that a lar
ge number of aircraft are being modeled. In many applications, predictions
of the number of times an aircraft can fly each day and the number of aircr
aft that are ready at any time are applied to a small fleet of aircraft. Th
is paper demonstrates the improvement in computational accuracy that is ach
ieved by reflecting the impact of small numbers of aircraft on availability
projections. The approach used is to extend existing finite queuing spares
models to including redundancy. Further, the method is used to optimize sp
ares provision with respect to a user specified availability goal. Although
the case study for this work is a military combat aircraft application fro
m the Gulf War, the method is applicable to any small system of vehicles or
machines where components may be redundant, demand and repairs may be appr
oximated as following an exponential distribution, and limited access to sp
are parts is the rule.