I am a big fan of
libraries. They are a fantastic resource for people interested
in Continuous Improvement
.I tend to always find myself in the non fiction section.
I don't know if you have come across the "Dummy" series of books. I think they
are great way to get some essential knowledge on a topic quickly.
There is even a book "6 Sigma for Dummies". Anyway I was at that stage where I
put down my books on the counter. You know that stage where the librarian is
secretly assessing what type of person you are based on your book choices. I'm
sure assistants at supermarkets do the same thing.
Anyway as I put lay down my five Dummy books on the counter there was an
embarrassing silence as the librarian scanned them in. I had to speak.
"I promise you I am not really a Dummy." I said.
It was funny in my own brain.
The librarian smiled with a knowing look that only librarians can have.
Anyway that is what prompted the title of this article. 6 Sigma for Simpletons.
I am not suggesting that you are a simpleton. Of course there is a chance you
may be, but because you have signed up for this very informative, entertaining
and modest newsletter then I think the chances of you not being a simpleton are
See what I did there. I nearly insulted you to get straight into the guts of
what six sigma is all about.
Where did the concept of
6 Sigma get popularised
Where did the concept of 6 Sigma get popularised and so what?
Six Sigma really came to life as a business improvement strategy at the
Motorola Corporation in the USA in 1981. To cut a long story short Motorola did
They reviewed all of their product defects. They then picked the defect that
was the biggest problem. They then looked at the process involved in creating
Their goal was to try and eliminate the defect by reducing the variation in the
process and all of the potential reasons for the defect. The idea was to bring
the defect level down to 3.4 defects per million. Another way of saying this is
that 99.99966% of all of the products produced are free of the defect.Wow! .Why
per million I hear you cry. 3.4 parts per million represents a 6 Sigma
So when you hear people talking about 6 Sigma, who do not really
have a clue what it is, (and there are a lot of them about) you now know
more than them.
So if a process is 6 Sigma then it only produces 3.4 defects per 1,000,000
Most organisations only have a 3 sigma process of 67,000 defects per million
transactions. (Stated even more simply this represents a 6.7% defect rate)Let's
go a bit deeper but not too deep
So, if you didn't know before, hopefully you now have got the idea that a 6
sigma process is one that is pretty good!
In the next little section I am going to attempt, without the aid of a safety
net, to explain to you the mechanics of six sigma in really simple terms,
without you having to becoming a Master Black Belt, own a program called Mini
Tab, and bore people at cocktail parties. Do people still have cocktail
Everyone knows what an
Everyone knows what an average is don't they. Of course they do. Remember
when you were at school and you measured everyone's height in the class .You
added up all the heights and then divided the total by the number of children
in the class. That was the averageWell in fancy pants statistical engineering
we call the average the mean.
Why? I don't know. It is a shorter word. It rhymes with Lean? There is actually
a reason but I will not go into it for the fear of making you sleep.
So let's go back to the classroom and think about all of the children that were
measured. There are a few really tall children. There are few really really
small children. Then there are a lot of children round about the same
That is what we call a normal distribution. If we were to plot out all of the
heights on a bit of graph paper it would look a bit like a bell shaped
The mean value would be in the middle of the bell shaped curve.(The diagram
next door is not an actual bell shaped curve, but it does have a similar type
So I hope you get the idea,and can imagine plotting the heights of the
You will need to sit up straight, rub your eyes and concentrate for the next
There is an exam at the end.
It could affect your promotion prospects.
So please pay attention and take notes because I'm going to go through the next
bit quite quickly, and I will not be going over it again.
The Standard Deviation or
Sigma (as in Six Sigma)
The Standard Deviation or Sigma (as in six sigma)
I'm not going to go into how to calculate the standard deviation. It is
basically the mean of a mean
It shows how much variation there is from the "average". A low standard
deviation indicates that the data points tend to be very close to the
A high standard deviation tends to indicate that the data points have a bigger
variation from the mean.
For instance say we gave 20 random car drivers the chance to race a formula one
car around a race track. I suspect the lap times would vary quite a bit.
If we then asked the top 20 formula one drivers in the world to go around the
same track then the variation in lap times would be very small. i.e. the
standard deviation would be very small.
Going back to our classroom example, let's say that the mean height of the
pupils was 140cms, with a standard deviation of around 8 cm.
A fair number of pupils, about 68%, will have a height of between 132cm -
148cm.That is one standard deviation from the mean.
The majority of pupils, 95% will have a height that falls within two standard
deviations (16cm) or two sigma from the mean. i.e. 124cm - 156cm.
Hope fully you are getting the idea...
So for a process to be 6 sigma 99.999976 % of the population need to fall
within six standard deviations from the mean!
So when we strive for a "6 Sigma" process we are looking to drive down the
standard deviation or sigma. How we do this is a topic for another day.
If you didn't before, I hope you now understand what "6 Sigma" is all about and
have found my explanation useful.
Hope this helps