I really do think the idea of Reliability Centred Maintenance is wonderful. However the huge corporations that run them for their maintenance scheduling have something we don't have: Lots of data.

As do Insurance based maintenance management companies, they have a large database of failure reports and can mathematically assess the risks involved in maintaining to any particular level. Unfortunately we can't.

Such information could probably be purchased but the cost would be prohibitive for any individual bay.

To gather your own data would presumably require one to apply maintenance on a reducing scale until things are seen to go wrong. Then risk assess whether to hold that level of maintenance, step it back up a notch, or continue the experiment.
The data collection over such small samples as an individual NHS trust owns would therefore I guess take place over a number of YEARS to be representative. Isn't that all rather time consuming in itself? Isn't the service bay in a rather exposed position in the interim?

It seems extraordinary that the NHS as a whole can't somehow utilise its own data and do itself a huge favour.

Marc