Reliability and maintainability allocation to minimize the total cost of ownership in a series-parallel system

Abstract

Allocation of system level requirements is most effective when performed early in the system's design phase. This holds especially true for two critical and fundamental design characteristics: reliability and maintainability. Traditional reliability allocation models are developed to either maximize system reliability under a cost constraint or minimize cost subject to a system-level, target reliability constraint. Cost, in these traditional allocation models, is represented solely by unit cost. Unit cost, by itself, is an inadequate measure of a system's operational effectiveness. In fact, the underlying economic metric used to properly describe the operational effectiveness of a system is total cost of ownership (TCO). TCO includes not only the upstream unit cost but the downstream operations, maintenance, and support costs. In this paper, new allocation models are developed based on TCO that simultaneously allocate both reliability and maintainability for a series-parallel system subject to meeting a system-level availability target. A non-linear representation of a mathematical model is defined that simultaneously allocates both system-level reliability and maintainability targets in a manner that minimizes TCO. This non-linear model is then transformed into a surrogate linear model that can be solved using existing commercial software. Examples are then discussed to illustrate the solution procedure and to show the sensitivity of allocation design decisions to fluctuations in economic factors such as discount rates, and design factors such as the life of the system.