What is the deal? In the beginning, the
Six Sigma beginning of course, I don't go that far back, there
wasn't any Six Sigma consulting companies. The only politics
that you got into was people in your own company that were
resistant to change and didn't want to change to Six Sigma.
Being resistant to change is human nature, but is a problem
even today with starting a Six Sigma or any other program. On
top of that, but we believe that the concepts have grown into
the proverbial 1000 pound gorilla.
It use to be, in 1985-1990, that Cp
was calculated like this (USL-LSL)/6sigma
actual. Now they have thrown in Pp
and PPk. Pp is calculated like this, I guess, (USL-LSL)/(rBar/d2)
by some but apparently not by all. The rBar/d2 is an estimate
that is, in part, used with range charts. Where in the world
did this start? Who said it and what research have they done?
I've been trying to find out but keep running into dead ends.
And I believe it was Besterfield in 1987 stated that Cp was (USL-LSL)/6sigma.
Regardless of what one may call it, the important measurement
is (USL-LSL)/6sigma and estimates are subject to error, so why
even use them.
What this really does is make things much more complex than
they have to be, which is contradictory to what I was taught
Six Sigma is about. Six Sigma is, or at least use to be, about
making things simpler and easier to understand. It was about
streamlining processes and not doing anything that was non-value
added and adding items that were value
added. It was about common sense. But now it could be a
perception that it is about making things as hard as possible
so that you have to hire a team of consultants at $1500.00 per
hour to explain things to you. In return, if you are really
lucky and you production is already so messed up that anything
would help, you will get a return on that investment in about
10 years. You should be getting a return on that investment in
half that time. Of course measurement is just one aspect of a
Six Sigma program. But it is the one most people have a
struggle understanding. And the more complex it can be made by
some people or organizations, the better for those people and
organizations being hired. Of course not all people and
organizations are that way.
For example, I recently got an ad from a Six Sigma
organization wanting me to advertise with them. They sent me a
courtesy copy of their newsletter. In the newsletter, they
were bragging about a conference they had and about the
speakers that spoke during the conference. Stating that one
speaker from a big company based in Illinois was the star
attraction. They stated that this company was a 'industry
leader' in Six Sigma. Well, I have talked to some of that big
company's suppliers and believe me, they are not an 'industry
leader' in Six Sigma, at least not in my opinion.
For openers, the large company's purchasing agents told their suppliers
that they would cut cost by 3%. Now this would be OK except
that the big company didn't have everyone in their company
trained in Six Sigma. They didn't have a vendor
council established so that suppliers could bring up problems,
usually more often than not due to politics. Without the
council in place, at that point in time, they should not have
been telling their suppliers anything except here is what we
are doing. When the suppliers ask them when the council would
be in place, they were just told soon. No time frame of when
it would be completed, but the suppliers was told, cut cost,
and ah, oh yea, you need a black belt, so train someone. The
big company also refused to commit on guaranteeing any amount
of business in the future.
It is asinine to expect a
supplier to do all that before a council is even established and
without committing to giving them enough future business to
justify the additional cost of getting 'tooled up'. And it's not
like the supplier really had a quality problem. A process
engineer told the supplier that compared to other suppliers,
they didn't have any quality problems. This is pure politics and
a gross abuse of power by the big company's purchasing agent. Of
course as we all well know, abuse by power hungry purchasing
agents is uncommon. This is not Six Sigma. This it is politics.
Now maybe this big company has since gotten it's 'act' together,
but, I bet that whatever person from whatever company this Six
Sigma organization could get to speak at their conference would
be a 'Six Sigma Leader". Once again, this is politics, and
politics costs, it doesn't pay. But only the 'Top Dogs ' can
make politics go away, and it seems to not be on their priority
list, so we must learn to live with it and deal with it in a
elegant manner, and bite a lot of bullets.
Here is something to think about. What would happen if all the
companies that are allowed to do 'self-audits' for ISO,
QS, or what ever, were not allowed to perform 'self-audits'?
What if they were not allowed to from the beginning? Would they
have jumped on the ISO and QS bandwagon in the first place? I
bet not. Now maybe I have seen 'Conspiracy Theory' once too
often, actually I've only seen it once, but let's think about
this.
An organization approaches a big company and states, we have a
concept to improve quality and we want you to endorse it.
Without the big companies endorsing the idea, smaller companies
almost certainly will not adopt it even if it improves their
profit margins. It is the concept that Quality Is A Necessary
Evil. Big company looks over specifications and says, 'OK' but
we aren't going to invest the thousands if not millions of
dollars it will take for us to comply. It will cost smaller
companies less to implement the ideas because they are smaller
and are less diversified. Even though in reality, the ratio of
overall dollars vs cost may be about the same.
So the big companies are allowed to do self-audits and not
required to invest the dollars and time into the concept. The
smaller companies on the other hand must in order to continue to
do business with the bigger company. The bigger company now has
a set of guidelines that their suppliers must follow. The big
company does not have to maintain or update these rules, and in
theory, their suppliers will have a better product. It is a
win-win situation for the big companies and the organizations,
but guess whom it is really costing?
Now don't get me wrong. If I were a big company, I would think
that it was a good idea too. And I am a firm believer that you
must have statistical
process control, (SPC), in your production process if
you want to reduce scrap, but you have to be able to get a
return on the investment. You must be able to see where that
return is going to come from. I see companies all the time, even
in today's world, trying
to inspect quality into their product. It can't be done. Let
me repeat that, YOU CANNOT INSPECT
QUALITY INTO YOUR PRODUCT! You are already to late by
the time you find a defect.
Six Sigma does not necessarily equal TCS, (Total Customer
Satisfaction). Sometimes it just means more profit or sales for
the company that has implemented a Six Sigma program correctly
and nothing to do with customer
requirements. Here is an observation that I have made over
the last three years.
GE is a Six Sigma company. That should mean that when their
light bulb packages say the bulb will last 1000 hours, there is
a very high likelihood that you will get no more than a few more
hours over 1000 hours as the package states. If it's less, they
will be in trouble with the U.S. Government, so companies
typically make sure that their products do at least what the
package states by a certain percentage. GE light bulbs usually
cost more than the no name generic brands, but because the no
name generic brands may not and probably do not use SPC, let
alone apply any Six Sigma concepts, you get many more hours from
the no name generic brands than from the GE bulbs. This is
because they still must guarantee at least the number of hours
that their package states, but they have less control over the
production process. Ergo, to insure the minimum hours they must
add 10, 20, or perhaps 30 percent more to the life of the bulb,
depending on their production line and materials, etc. whereas
GE may only have to add 5 percent or less because of having a
very tightly controlled process.
Even if the no name generic brands used just a standard SPC
program with a 3 sigma design margin, it is nowhere near as
tight as the 4.5 sigma design margin, but it would be much more
cost effective than no control at all because they may sell more
units.
The last time I was in Thomas Edition's summer home in Ft. Myers
FL, it was about 1972, there were still light bulbs burning that
was made my Edition. So it is technically possible to make
standard incandescent light bulbs to burn for about 100 years,
if they did though, you would only have to replace a bulb if you
broke it. This would make a light bulb cost about $100 each, so
there has to be a wear out some place. While Edison's bulbs were
not very bright, they were still glowing. I reckon it has
something to do with the tensile strength or thickness of the
filament or the inert gas. But, I was changing GE light bulbs it
seemed like every 6 weeks or less, and of course they always
stopped working at night, when the stores were not open and when
you need them the most.
So I now only have 3 standard incandescent light bulbs. One bulb
over the stove, one bulb in the hallway, and one bulb in the
bathroom. I switched to fluorescent bulbs everywhere else. I
have had to change only one fluorescent bulb in 3 years, and I
am saving energy too. It was a real good Six Sigma solution to
the problem, and I didn't even have to spend thousands of
dollars to do it.
Update: It
seems that at some point after this article was published, GE
changed their light bulb boxes to indicate that the 1000 hours
is based on average usage defined as 2 hours per day. Apparently
someone at GE decided that 2 hours was an average.
According to the A.C.
Nielsen Co., the average American watches more than 4 hours of
TV each day. I'm assuming that the majority of the population of
this average is based on more nighttime viewing than daytime due
to most people, at least when the survey was taken, had daytime
jobs. No doubt GE knows this. I have my kitchen light on for
more than 2 hours a day, on average. I do cook, bachelors may
have a lighter usage load, but I don't see a family using less.
The ONLY room I may use a light for 2 hours in, on average, is
the bedroom.
So by using an unrealistic
average, they could degrade the reliability
of the filament, thus reducing costs, and achieve six sigma
goal. Wonder if this occurred when Jack Welch was in charge?
