Life cycle assessment of system availability and cash flow by stochastic modeling and simulation

Availability is an important performance requirement for system designs. At the design stage the availability statistics must be predicted over the life cycle span of the system. The life cycle is a probabilistic presentation of events and time of occurrence of the event. The author deals with one single category of events, namely shutdowns due to faults, followed by subsequent repair and start-up. Closely tied to the availability is the cash flow of the system. Availability and cash flow are strongly correlated, but each has its own statistics. The usual approach to evaluate system availability and cash flow by stochastic simulations is demonstrated. Those are based on a priori assumptions of probability density distributions for shutdown, repair and start-up. The central limit theorem shows that the variance in the availability is roughly inversely proportional to the life cycle span, which becomes a practical rule of thumb that applies to all property distributions. The net present value estimate of the cash flow has a higher variance than the others; this is due to the discounting factor