Optimum Replacement for CFM

[Robbie]

If I want to model optimum replacement time for a module which has a number of different potential items/components that can fail, should I model the system using the CFM in Weibull or use BlockSim?

I wondered if I should analyse the items individually and select the one with the shortest replacement time. In this case, having one item that fails means the complete unit is pulled and all major components refurbished, as its a electro/hydraulic/mechanical control module that sits on the seabed in a water depth 600ft. We may have more than 4 competing failure modes, but Weibull++ looks like it only caters for 4.

I tried doing it in BlockSim, by making a block with a sub-diagram, however I was sked to allocate approx. failure distribution for the diagram, which is made up of several items. Any thoughts on how to do this accurately? Thanks.

[Pantelis]

Well the question is what are you going to replace? If the whole thing is an LRU then creating (aprox) a failure distribution as you said is the right thing to do. If you are going to replace items inside then you should look at these items individually.

[Robbie]

We'd replace the module with a new or refurbished unit, and send it back for refurbishment (regardless of which failure mode occurred). I guess my question is, how do I assign a failure distribution for the system as is required in blocksim. The approximate failure distribution opens a box which requests distribution settings that are not given in Blocksim when a model is run - or perhaps it does, I just don't know how to do that yet.

[Pantelis]

Use that .., it will generate data and put it into Weibull++ ...

[Robbie]

Worked it out now, thanks. This is a great function for some current work I have - the more I use BlockSim and Weibull++, the more impressed I am.

[Pantelis]

Good to hear