Page 408 - Improving Machinery Reliability

372 Improving Machinery Reliability

made it their goal to eliminate maintenance as it was then known, the benefits were
drastic and immediate.
This company now recognized something to which we had alluded before: The
reliability workforce members must be made up of well-motivated self-starters-
men and women with inquisitive minds. They must be supported and valued by man-
agement, since no self-respecting reliability professional will be happy and produc-
tive in a stifling, business-as-usual work environment. In addition, this reliability
workforce will only become optimally effective once the reliability statistics at the
local plant site are being compiled and are routinely made available for comparison
to representative or equivalent industry statistics. This is commonly known as bench-
marking and implies that data collection is taken seriously by every job function in
the plant.

Benchmarking: Comparing Yourself to Your Competition

If you wanted to know the extent to which it is reasonable to make improvements,
you would compare your maintenance performance against that of the competition.
Numerous different definitions and benchmarking routines exist and it is outside the
scope of this segment to debate their merits and shortcomings. However, it should be
intuitively evident that only a meaningful definition of availability will do; a random-
ly chosen availability statistic alone will not tell the full story. Say, for instance, an
automobile that is being serviced twice a year will be unavailable for two days; hence,
it has an availability of (365 minus 2)/(365), or 99.45%. Another car might suffer
from an electronic glitch that randomly shuts it down once per day and for just one
minute per event. It is thus unavailable for “only” 365 minutes each year. Since 360
minutes would be six hours, or one-fourth of a day, the availability claim of this car
could be a seemingly attractive value of 364.751365, which equals 99.93%. And yet,
would not a reasonable person prefer to own the less available car?
We found the information given earlier in Chapter 4 (Reference 5) helpful; several
tables in that chapter give strategic level measurements. We consider them rather
self-explanatory, but wish to direct particular attention to the fact that in 1996, the
certified training costs in best-of-class companies were $1,200, while the “average”
plant spent a disappointing $400 per employee. Add to this that much of the so-
called training does not necessarily impart real, useful, or implementable knowledge.
And now draw your own conclusions as to the state of training and up-to-date tech-
nical competence at some plants that talk reliability but fall far short of pursuing the
most cost-effective implementation steps.
Appendix A contains relevant statistics, some compiled over years of observation
and data taking. Please note that some performance measurements are expressed as
reliability, availability, duration of downtime, failures-per-million operating hours,
mean time between failures (MTBF), or mean time to repair (MTTR). Knowing
where you stand is without doubt an important prerequisite to either the “designing-
out” of maintenance or to the adoption of RCM. Among other things, it will tell you
whether your equipment is close to being life cycle cost optimized. Do you have
these comparison data? If not, how will you get them?