Hard time and soft time maintenance

[MarkPlett]

I would like to model a life-limited part (LLP). Its reliability is essentially 100% until the end of its service life. When a module with an LLP installed is operating it can fail for various reasons. When the module fails, the inspector checks the service life of the installed LLP. If it exceeds 25,000 hr, the LLP must be replaced with a new one. If the module doesn't fail but operates until the service life of the installed LLP reaches 30,000 hr, the module must be taken out of service to replace the LLP. Two possible fail situations are:

LLP reaches service life of 30,000 hr (hard time), or
Module fails and LLP exceeds service life of 25,000 hr

[Arai.M]

Mark,
What are the specific answers that you are trying to get out of the model?
Arai

[Arai.M]

Mark,
To expand on my previous reply...if what you are looking for is probabilities of one situation vs. the other one, you could model that in BlockSim as follows:
Have blocks in series configuration, one (or more) block(s) for the module's failure mode(s), and another one to represent the LLP. Now the LLP would have a reliability set as normal with mean 30,000 and a small std. dev. describing the hard maintenance as a deterministic event. Additionally, define inspections for the LLP with the default policy, which is "on system down", so that every time a failure occurs for the module, the LLP will be inspected. On the inspection tab, there is something called Failure Detect Threshold which should be set to 0.8333 (25,000/30,000). This will trigger a preventive maintenance (your soft maintenance) every time an inspection finds the LLP has exceeded 25,000. Make sure you define a PM for the LLP as well. Your results should contain among other results the number of failures (hard maintenances) for the LLP as well as the number of PMs (soft maintenances) that occurred during the mission time.
Hope this helps,
Arai