Benchmarks #3

Mean Time To Repair (MTTR)

Repair planning and repair management, are key factors in a another performance measure known (in engineering terms) as “mean time to repair” or MTTR. Mean time to repair is a measure of how slowly (or quickly) a machine is returned to service once a downtime incident occurs. MTTR is becoming one of the benchmarks that many mines are using to evaluate their performance – particularly repair or maintenance efficiency.

Efficiency -- a measure of turnaround time or efficient use of downtime in getting a machine back into service – expressed as Mean Time To Repair (MTTR).

Repair turnaround (the MTTR number) has an impact upon mechanical availability but the impact is not as pronounced as that from MTBS. Graph #3 (above) shows that as MTTR (turnaround time) increases, availability decreases. The change is close to a straight line so any incremental change in MTTR has roughly the same impact upon mechanical availability.

MTTR is a relatively direct measure of efficiency in both planning and execution. It is calculated by the following method:

MTTR = Downtime Hours / Number Of Downtime Incidents

Delays (mechanical downtime without active work) are included in MTTR.

Many factors can effect MTTR. A proactive (as opposed to reactive) management philosophy is one of the most important. Timely information and proper planning based upon that information are also important factors in getting a machine back into service in the shortest possible time. The following items illustrate how MTTR is effected.

¨ Decision-Making / Decision Timing – deciding what course of action to take in completing repairs can significantly impact repair times. Incorrect decisions are often the result of incomplete information or improper planning. Timing is also critically important. If repair planning, and the ordering of needed parts, is not started before the machine is in the workshop, downtime will be unnecessarily increased.

¨ Parts Availability – the lack of readily available (and on-hand) parts can result in long delays in the work bay. Unusual or unexpected parts requirements may be unavoidable. However, the majority of parts need’s can be “anticipated” (using condition monitoring) and planned for.

¨ Tools – if the proper tools are not available, or are being used elsewhere, repair times will be increased. Having both the correct tools and correct numbers of tools has to be considered.

¨ Equipment – the lack of specialized equipment can cause delays or reduce efficiency. This would include equipment for lifting, blocking, cleaning, component handling, fluids’ delivery and removal.

¨ Available Bay Space – if facility burden is too great, work must be delayed or done under circumstances where efficiency will suffer.

¨ Available Manpower – if skilled manpower is unavailable, repairs may be delayed, done inefficiently or done by unqualified workers.

Because of the importance of the above items, many systems track, keep records and analyze delays that occur when the machine is down. (These records often show total machine downtime and the breakdown of that total time to show the reasons for delay in detail; i.e., waiting for parts, waiting for shop equipment, etc.).

Many mines have begun to measure MTTR and to use it as a benchmark. MTTR can be used alone and it can to help analyze mechanical availability.

When used alone, MTTR is graphed over time (monthly for example) to establish a trend in repair turnaround (efficiency). Graph #4 (above) shows an example of how a change in repair turnaround will look when graphed. This graph and the trend line shows that repair turnaround time is increasing. This could be the result of increasing inefficiency or it could be caused by the nature of the work being done; e.g., long duration jobs such as component changeouts.

The examples given above illustrate how MTBS and MTTR can be graphed alone to identify undesirable variations or trends. These two factors can also used to analyze mechanical availability. The relationship between MTTR, MTBS and Mechanical Availability is shown in the next section.