By Steve Turner of OMCS Pty Ltd

Paper first presented at the International Conference of Maintenance Societies, Melbourne 2001

In part 1 Steve explains the basics regarding a effective maintenance management strategy called Planned Maintenance Optimization (PMO).  In part 2 we learned how PMO compares with Reliability Centered Maintenance (RCM). This final segment explains some of the statistical tools that are used to perform analysis and how they compare with real world experience.

There are two groups of approaches to determining task frequencies. One of them is statistical in nature and the other is more empirical. There are significant differences between the two approaches and some of these are discussed in the following paragraphs.

Mil STD 2173 (AS)

Mil STD 2173 (AS) shows that for random failures¹ , the optimal Condition Monitoring frequency is given by the following formula:

Where

•  T = Age (time) between the point at which the failure can be first detected and the actual failure - also known as PF interval.

•   n = Number of inspections during the PF interval T.

•   MTBF = Mean Time Between Failure

•   Ci = Cost of one inspection task

•   Cpf = Costs of correcting one potential failure

•   Cnpm = Cost of not doing preventive maintenance inclusive of the     operational costs of failure.

•    S = Probability of detecting the failure in one inspection (task effectiveness).

¹ A similar but more complex formula can be applied where the failure pattern is not random.

Click here for page 2