Inspection upon System Down

[Jennifer.Bell]

I have two questions relating to the "system down" state and inspections.

System set up in the following manner:

- System set up with identical blocks in parallel.

- Block failure causes corrective maintenance. Although system is not set up in series, corrective maintenance causes a system down.

- Repair crew for corrective maintenance has a fairly long logistic delay and repair time.

- Non-failed components continue to run (age) upon system down
- Inspections occur for all running components. If within a certain time of failure, will trigger a maintenance action, otherwise component will not fail.

I have two questions.
1) When is the timing of the inspection? Is it as soon as the crew is called, or after the logistic delay?
2) I need to count all failures in a time interval, hence the CM causing system down. Is there a way to (a) count failures by block by time interval? I have seen only total simulation results, or (b) allow throughput to continue through non-failed blocks even while system down?

Thanks

Jennifer Bell

[Arai.M]

Hello Jennifer,
1) Inspection downtime is assumed to start as soon as the crew is called. The duration of the inspection itself starts as soon as the crew arrives. Therefore the total downtime for the inspection would be the sum of the crew logistic delay plus the duration of the inspection.

2) If you select to bring the system down with CMs and that is not what is happening, it will most likely skew your results and make it difficult to interpret them.

a) Results are reported at the end of the simulation. However, it is possible to use phases to obtain results by time interval. For example, say that you are running a simulation for a year but are interested in quarterly results. You would create a phase diagram with 4 phases, all pointing to the same diagram you have been running. Each phase would have a duration of 3 months. The results reported are mostly the same as the ones reported for standard simulation, however, a drop down will allow you to navigate results by phase as well. For more information on phases, please refer to http://www.weibull.com/SystemRelWeb/reliability_phase_diagrams_(rbds).htm.
I have also attached a quick example that might help you get started.

b) if the block is set to operate through system failure (from the general tab in the block properties, check Item operates even if system is down) then the non-failed blocks should continue to operate. Keep in mind that if they are all getting an inspection, they won't be operating unless you select to perform a non-downing inspection.

[Jennifer.Bell]

Thanks for your quick response.

on system downs - still wrangling with how to get the reported results I need - but that is a very helpful tip on using phase diagrams.

One final question on inspections. If a different crew is in charge of inspections, which has no logistic time delay, will the inspection still begin when the corrective maintenance crew arrives?

my goal here is to simulate the scenario that if a failure is detected while another failure is being repaired in the system, the repair times are slightly different - the crew does not require a logistic delay. I'm hoping inspection will allow me to do this.

Thanks,
Jennifer

[Arai.M]

Jennifer,
The inspection may have a different crew from the corrective maintenance in which case, the delay of the corrective maintenance crew will not apply.
Keep in mind that in BlockSim, if an inspection finds the components near failure, a preventive maintenance will be triggered (not a corrective maintenance). This should allow you to capture the differences better.
For an example on how to setup on-condition maintenance, see http://www.weibull.com/hotwire/issue62/hottopics62.htm