Reliability Performance Indicators - Pressure Safety Valves

[novice123]

What should be the reliability performance indicators for pressure safety valves?
How the testing data should be assessed to get meaningful indicators which should further be analyzed?
One valve failing in pop, other in reseat, another with backpressure test failed.

[octane]

Failing safety valves should be investigated on case to case basis and no failed case should ideally be missed as that becomes an opportunity to avoid any process upset before it happens. Every failed safety valve calibration test is an indication that in case of an actual pressure demand, there would not be having a lift as per design leading to incidents with consequences.
Pressure safety valves failures vary typically from their popping to their seat leak performance, and during the tests, there are the indicators which are recorded on the calibration certificates. Further, depending on the type of pressure safety valve, backpressure test, pressure vacuum relief valve, another vacuum relief testing comes in.
I would suggest you to have a working data on the management of PSVs based upon their type and the tests performed. Digitize the inputs to know any failing tests, their repeat cases, common contributing causes /factors, and then run analysis on Quarterly basis supported with a report and a list of bad actors. Those bad actors are then worked upon to perform root cause analysis.

[tuan]

This is one area which needs closer focus than usually it gets.
Any PSV found failed in the bench test should be reported as a near miss. And that particular near miss, as related to plant integrity, should be investigated.

[irish]

See just go through these indicators and I am certain you would know what kind of base data is required to be maintained to perform a reliability analysis focusing on PSVs failures:

As-found pass rate
As-found failure rate
Dangerous failure rate
Set pressure drift
Leakage rate
Failure mode distribution
Mean time between failures
Failure rate by service
Failure rate by manufacturer
Failure rate by PSV type
Reliability versus test interval (eventually you want this)

These 11 indicators would allow you to point out where there is a problem and you may initiate a formal RCA for identified cases.

[mechcolor]

I would suggest that for some of the PSVs having repeated failures, manufacturer should be involved.
We had a case where an installed internal for PSV was identified as having got aged after 30 years of service.
Upon asking from the manufacturer that where that has been mentioned, he replied that we were not expecting that a regular inspection would not identify an issue triggering a replacement. And yes, we were performing inspection for that particular internal but never thoroughly knew its failure methods.