Forecasting MTBUR based on past performance

Hi there!

I came across a scenario recently that I think the logistics team I am interning with is approaching the problem incorrectly, but I am not sure; I was wondering if you could check it out for me.

The Mean Time Between Unscheduled Removals (MTBUR) of a certain part is say 5,000 hours. There is a service contract on the part for 15,000 hours. Therefore, the forecast demand for the year is 3 parts to be repaired. 4 months into the contract term, 5,000 hours have gone by but no part has been removed yet for repair. Now, to forecast the demand of repairs for the part, what do I do?

The previous method that the Logistics Team here uses is that they take the original contract hours (15,000), subtract whatever time has gone by (5,000) to get 10,000, and then divide by the MTBUR of that part (5,000) to get 2 repairs. But I don't think this is correct, because we are ignoring the fact that the part has already been in use for 5,000.

Let me know what you think and I look forward to hearing back! Thanks for taking a look!!

Always my best,