Cost of increased reliability

[Jørn Hennum]

I am doing an initial RAMS study of a radio telescope array that is still in the conceptual phase. I would however like to be able to say something about the cost of increasing component reliability, which is hard because there are no data available whatsoever, and also no prototypes are available for testing. Therefore, I wondered if anyone on this forum knows if there are any generic models that at least say something about cost trends as reliability increases. Needless to say, higher reliability inevitably costs more money, but how much, and are there any models available to address this problem?

[RS Support]

As you said increasing reliability would increase initial costs (however it will reduce subsequent costs – i.e. downtime, maintenance, warranty etc.), thus there is an optimum where the sum of these costs is a minimum.

Now, on your question as to how much, well this is fairly broad and it depends on many factors, thus even a generic model would require some quantitative inputs form you.

A simple model used to allow for the inclusion of generic costs in reliability allocations schemes is discussed at http://www.weibull.com/SystemRelWeb/reliability_allocation_and_optimization.htm

[Tobbe]

The cost of reliability must be compared to the "consequence of failure". It's as simple as that. It's of course hard to estimate what the cost of a component, or a system not working. You'll have to step into the accounting word and do what they do.... make qualified guesses and put numbers on each scenario for what may happen if it fails.

There are many practical and good articlea at www.idcon.com

[Larry Strouse]

Does BlockSim support generating a graph of cost per unit time vs. replacement time as shown on page 217 of the BlockSim 6 system analysis handbook?

[Pantelis]

Larry, not directly. You can generate the values in Weibull++, from a general spreadsheet, using the Function Wizard and the "Table of Block Replacement Times" function.