Optimal designs of (k,n-k+1)-out-of-n:F systems (subject to 2 failure modes)

The problem of achieving optimal system size (n) for (k,n-k+1)-out-of-n systems, assuming that failure may take either of two forms, is studied. It is assumed that components are independently identically distributed (i.i.d.) and that the two kinds of system failures can have different costs. The optimal k or n that maximizes mean system-profit is determined, and the effect of system parameters on the optimal k or n is studied. It is shown that there does not exist a pair (k,n) maximizing the mean system-profit.<
>