The authors try to answer the question, “Are the methods of M (maintainability) demonstration sufficiently developed at this date to make M incentive contracting practical?”
To answer the question the authors review and critique the M demonstration methods in vogue today. Seriously questioned are the present statistical techniques for reducing M demonstration test data.
The authors agree that sufficient field maintenance data exist to confirm that field times-to-repair of electronic and electromechanical equipment are best fitted by a log-normal curve; but point out that from their experience one can expect the times-to-repair observed during an M demonstration test to be normally distributed. Principally for this reason, the authors suggest that the Md TTR (median-time-to-repair) is a more suitable M index than the MTTR (mean-time-to-repair) for contracting purposes.
The point is made that variations in the observed Md TTR will exist from one demonstration test to another and from the demonstration test to the field experience, but that these variations are more likely due to human factors than to equipment factors. Four causes for these variations are traced to differences in repairmen skills, experience, learning and motivation. To cope with some of the variations, some coarse correction factors for the demonstration test data are given even though the authors admit that much more study and development are necessary to achieve the accuracy required for making M incentive contracting practical.