A new framework is developed that analytically links an engineering concept, a system's reliability, with both managerial and financial concepts, the system's revenue generation capability, and its present value. A communications satellite is used as an example to illustrate the use and insights that can be generated from this framework. For instance, after the development of a revenue model for a communications satellite, the cost of unreliability was quantified: the present value penalty for the lack of 100% payload reliability. Next, the value of redundancy was also analytically captured: the satellite incremental present value provided by payload redundancy. The central finding is that present value calculations of a technical revenue-generating system, in this instance, a communications satellite, that do not account for system's reliability overestimate the system's present value and, thus, can lead to flawed investment decisions. Finally, when sensitivity analysis is performed on the various drivers of the satellite present value, it was found, against conventional wisdom, that redundancy in communications satellite payload is overrated; in other words, that increasing payload redundancy provides limited incremental value to the spacecraft.
Nomenclature
F(t) = probability distribution function that item fails within time interval [0; t] L(t)= satellite load factor as function of time, number of transponders in use at time t divided by total number of transponders onboard spacecraft; also known as utilization factor L 0 = average load factor of all current geostationary orbit communications satellites, approximately 60% in 2003 P i = price of transponder i per unit time P T x (t) = average price of transponder onboard spacecraft per unit time R n = reliability of item at time n T R(t) = reliability of item r T = discount rate adjusted for time period T T f = random variable, time to failure of item T x total = total number of transponders onboard spacecraft t wear-out = beginning of wear-out period of item u i = expected revenues generated between (i − 1) T and i T u(t) = expected revenue model of spacecraft per unit time, or system's utility rate after it reaches operational capability T = time period, small enough over which it is appropriate to consider u(t) and λ(t) constant λ(t) = failure rate of item τ = fill rate time constant