Definition:
Mean Time to Restore (MTTR) (also can be called mean time to restoration) is the average time required to detect, repair and restore a system to operational condition after failure. Very similar to Mean Repair Time (MRT), but different. MTTR is only utilized for SIFs that have diagnostics. This is an often confusing concept. It is best to break it down in the four stages:
a – time to detect the failure with diagnostics
b – time spent before starting the repair
c – effective time to repair
d – time after repair to put back in service.
MTTR = a + b + c + d –> from the failure itself to the restoration
MRT = b + c + d –> meaning MRT does not include time to detect.
MTTR = a + MRT
MDT = a –> mean detection time is equivalent to this
MTTR impacts PFDavg and STR calculations. It impacts PFDavg calculations only if a detected failure notifies only and does not vote to trip. The STR is impacted if a detected failure votes to trip. The nuances of this are a bit complex and are best for another post.
See this Blog Post to see more details how MTTR impacts your PFDavg and STR calculations.
Key Points:
- Critical input for system availability analysis.
- Affects operational downtime.
- Note – this used to be “repair” and not “restore,” but is no longer correct
- For SIFs with modern diagnostics, this could make “a” near zero. This would mean that MTTR can be essentially MRT. But this may not be true with older instruments with diagnostics as the diagnostics did not run continuously
Example:
A plant has a SIF with 2oo3 logic on the instrumentation. The facility is willing to spend the money to keep 2 instruments in stock at all times along with Technicians who always on watch. It would be reasonable to have an MTTR of 8 hours. If they were not willing to have the units in stock, and the local vendor representative does not either, then 2 weeks or more may be needed.
See also: MRT, MDT, DTI, MTTR Blog Post
Cited Source:
- IEC 61508-4:2010.