Quantifying the Business Case for Change must be the first step in any new process improvement initiative.

1. Appendix 1: Many rules of thumb exist that
suggest improving your lubrication
Quantifying the program would garner a 10%
Business Opportunity
reduction in maintenance costs,
or that implementing RCM would
improve productivity by 20%, or that
he very first challenge with
T
implementing precision alignment
any improvement initiative is standards would decrease energy
to demonstrate that spending consumption by 7-11%. While all of
money to improve something will these have been true for scores of
actually yield a return. Essentially, plants, when it comes to our plant,
we must build a business case we are skeptical about attaching our
to show that the investment in name to a business case that is built
the initiative is worthwhile. While on rules of thumb, no matter how
most people understand this, the historically reliable they may be. So,
challenge comes in the next step: what is the answer?
agreeing on performance standards.
Business cases built on actual
data versus rules of thumb are
usually much more successful in
garnering support. Then enters the
next challenge, and that is one of
design parameters. Immediately,
people begin to argue about whether
a machine could do this or should
do that, and whether or not it was
designed for this rate, or remember
that one day for about an hour it had
actually done twice that, and so on.
32 GPAllied, © 2011

2. Debates rage on about design scientific exercise, based on actual
capacity versus actual capacity data and void of any opinion
and quickly turn into philosophical or conjecture. The findings are
debates about unrealistic clear, and the hidden capacity of
expectations versus theoretical the production line is simply and
design and actual performance. quickly calculated, thus making the
These debates are further monetization of process losses easy
aggravated by the ever-present and painless.
“well, when I ran these kinds of
plants, I routinely got 30% more out The message is based on
of it than you are” type of comments. demonstrating the business value of
When the dust has settled, we are employing science and engineering
no closer to an agreed upon process as a process improvement tool
maximum than when we started the that can visually bring the value
debate. proposition we offer to life. These
plots allow us to provide a clear
This is where Process Weibull Plots vision of gains a process line can
can help. Immediately, 99% of make, and only requires an hour
those conversations become moot. or so of our effort using one year’s
Why? Because Process Weibull worth of daily production data in
Plots simply take a look at a plant’s order to do so.
actual daily production, find the
repeatable patterns of demonstrated
production capacity, and show the
effects of removing variation from
the process and holding that level
of performance for a longer period
of time. Thus, the opportunity case
for improving the performance of
a given line becomes a strictly
www.gpallied.com 33

3. What Is a Process Weibull Plot? This data is valuable for running
Process Weibull Plots, or more day-to-day operations, but it is not
simply process plots, represent a extremely valuable when it comes
new way of looking at production to understanding the value of gains
data. In most cases, industries look from improving process control and
at their daily production data in a reliability levels. Process plots give
time plot format, as shown below. us that capability.
In other words, they show production The process plot provides a wealth
data from day-to-day and calculate of information at a quick glance.
averages, standard deviations, With this plot, we are able to see the
and in some sophisticated cases, line’s normal level of process control
they even generate Statistical and the point at which process
Process Control (SPC) values for reliability is lost.
understanding when they may
be losing control of their process.
34 GPAllied, © 2011

4. The Beta value of the plot is the
Beta Value Control Level
prime indicator of the level of control.
The point at which the plot strays 5 Poor
from the production line is where
process reliability is lost. We can 10 Fair
then calculate the production gains 25 Tight
that can be made from improving
process and asset reliability, as 50 Excellent
well as the gains that can be made 100 World Class (Potential Six-Sigma)
from improving overall control of the
process. 200 Seldom Achieved
www.gpallied.com 35

5. How Would We Use This Line Monitoring and Performance
Information? Improvement
Performance is a function of three The Special and Common Cause
things: 1) your rate; 2) how much Losses can be calculated and used
variation you have in that rate; to benchmark a process line.
and 3) how long you were able to Special cause losses are those
maintain that rate within a given level losses associated with significant
of variation. It takes all three factors amounts of downtime. These
to attain high levels of performance. losses are very often the result of
Let any one of the three drop and commercial downtime, equipment
line performance is significantly failures, problems with raw
affected. materials, and/or supply chain
issues. Common Cause Losses are
Process plots are excellent ways to a lot harder to track and, when not
measure all three, and the data from using process plots, very difficult to
the analysis of these plots can be quantify. Common Cause Losses
used in numerous ways. are those losses that are usually the
result of short process interruptions,
equipment changeovers, variations
in raw materials that result in jams,
variations in operators, etc.
36 GPAllied, © 2011

6. Hidden Plant Capacity Calculations In summary, using a production
The losses associated with a line line’s actual daily output to calculate
equate to production units lost. line performance characteristics
Recovering those lost units in the and the amount of hidden capacity
future means more production is fairly straightforward utilizing
from the same amount of invested Process Weibull Plots. These plots
capital. The losses can add up to relieve the analyst of the burden of
numbers so large that capital dollars the traditional engineering versus
need not be spent. Meaning, if actual performance set points and
enough lines have enough hidden enable a useful and believable
capacity, then additional lines need calculation of line performance.
not be built if the business is in a
sold-out state. If the business is How would you shift Beta and
not capacity constrained, then two increase Process Reliability?
shift operations could be reduced
to single shift operations, provided
enough of the hidden plant capacity
is uncovered.
www.gpallied.com 37