# Defining and Calculating Sigma Level

Video Activity

Join over 3 million cybersecurity professionals advancing their career

Sign up with

Required fields are marked with an *

or

Already have an account? Sign In »

Video Transcription

00:01

Hi, guys. Welcome to Episode five Defining in calculating Sigma level I'm Katherine MK Iver and today will hopefully be more of a review for you where we're going to understand how to calculate d PMO and how to find a signal level using DPM Oh, and sigma table.

00:17

So if you think back to yellow belt remember, there's also 1/3 way of determining signal level where

00:23

a second way of defense determining signal level in yield rate. But I believe I talked about a little bit and yellow boat that I'm not crazy about yield rate because it can be very complicated because you have to reincorporate your rework eso. How many times did you do something?

00:41

Fix it and then it became acceptable to leave. I prefer DPM Oh, because I consider it to be more straight forward, way

00:49

of calculating defects and thus easier to use as a conversation point. So the first thing that we need to remember is why is it six sigma? So when we talk about six Sigma, what we're talking about is sigma is number of standard deviations from the mean generally we as

01:08

and culture.

01:11

So we photojournalism, we research scientists tend to live in the three plus or minus signals from the mean plus or minus three signals from the mean. So this talks about is, is what you're looking at right here

01:23

as a normal distribution curve or what we call a bell shaped curve. Um, if you got tired of downtime and yellow belt and the different ways you can identify waste, you are definitely going to get tired of the normal distribution curve in Greenbelt because this is really the greenbelt. Get down is

01:41

the normal distribution curves, so you're gonna be able to talk about this in your sleep.

01:45

It's really great at cocktail parties. If you want to avoid people, is start talking about how to read these. So

01:52

a normal distribution curve primarily when we're talking about a research application, we talk about plus or minus three standard deviations from the mean. So what that means is we have 100 data points. We can say that 99.73% of them are will fall within three standard deviations of the mean.

02:08

So it gives us an idea of predicting what's that likelihood

02:13

that something is going to be outside of that.

02:15

But because we are super cool and we are world class organizations, when we function at a six Sigma level, we're talking about plus or minus six standard deviation from the means, or we are in 99.99999 So there's five of them. Eight,

02:35

um

02:36

percent defect free or conversely, which is why D PMO is kind of my favorite is 3.4 defects per 1,000,000 opportunities. So if you have a 1,000,000 chances, 3.4 of them will have defects and you'll be functioning at a six signal level.

02:52

So what this means for an actual application? There are two industries that tend to actually really function

02:59

in the Six Sigma land, and the 1st 1 is aviation, so airlines. And if you think about all of the components that go into it, we have air traffic control and airport management and booking tickets and food, airplane food and baggage claims and those sorts of things. If you think about all of the different ways that you can have defects,

03:17

the aviation industry tends to function in a very high five

03:23

six. I've even heard some people argue. Seven signal level for all of the different ways that you can have defects. And my argument to everyone out there who has lost the luggage is think about all of the other places that you didn't have a defect. When you start thinking about the one,

03:39

the other industry that tends to actually function very, very high. Five

03:45

closer to six Sigma level is actually credit card processing. So with the exception of uncontrollable or special cause variation hacking those sorts of things credit card processing, tens toe have very, very low defect rate. So if you think about the number of times that you use your credit card

04:04

versus the number of times there has been a defect or in error on A,

04:08

they tend to function in the very, very high six or a very, very high five. Lucic signal level with the exception, of course, of hacking and external factors. But we're gonna talk about D, P, m o or defects per 1,000,000 opportunities.

04:25

So when we talk about are the number of times that you feel a defect, we talk about aviation and everybody out there like Whoa, whoa,

04:33

they lost my luggage when I went to Cancun. That's actually the reason why I prefer to use DPM. Oh, so we can talk about the different ways. They're the number of different ways that could have gone wrong and the number of different ways it actually went wrong. So DPM o stands for defects per 1,000,000 opportunities.

04:50

And to calculate this, you first need to find how many opportunities you have.

04:56

So in Yellow Belt, we used to the example of the form that has, like, 12 different fields. So there's a 12 different opportunities for defects, and then you count the number of defects that actually happened. So if my form has 12 fields, um or let's make the math easy and say My form had 15 fields

05:14

and there were four defects, I then would have a deep PMO rate of,

05:18

um

05:19

to those are 266. Yeah, 266,000 DPM. Oh, but so you take the the different with number of different ways with a number of opportunities for defect, and then you're going to divide the number of defects of by that,

05:35

multiply it by a 1,000,000 because it is per 1,000,000 opportunities.

05:40

That's how you get DPM. Oh, I personally really enjoy doing this exercise with higher level leadership and organizations because the higher the level of leadership you get, the less likely you are to hear all of the good things. You tend to only hear about the hiccups or the defects with the problems.

05:58

So for me, it's always really eye opening toe. Have our executives go through

06:01

a de PMO exercise so they can see that there are so many different ways that we're doing things right and you just don't hear about it? That's tends to be why I prefer to you PMO over yield re in addition to yield rate can be complicated to calculate when we re work When we had the reworking but so summary on DPM Oh,

06:21

count the number of opportunities

06:24

and then you're going to count the number of defects, do some quick division, and that's going to give you a D PMO ***.

06:30

Then we're going to look for our signal levels. So if for some reason you work in an organization that wants to report that they function in signal levels, well, you're gonna want to do is find a signal table. If you've taken a basic stats class. This will be one of the back pages or the appendices

06:46

in your statistics book, or it is very widely available to different levels of specificity. So

06:53

the example that I gave you is just the high level one significant figure signal levels. I've seen them go out to two or three where we're talking like 2.355

07:04

to give you that specificity. That depends on you and your comfort level. Generally speaking, because I live in DPM Oland, I tend to only want one significant figure, maybe two. So it's like 1.1

07:19

1.5 that kind of frame to give a signal level. So when you use a sigma table, you are going to find your DPM. Oh, and you're going to look for the corresponding Sigma table. If for some reason that you decided that you like to do things the complicated way and you do a signal or a yield rate,

07:39

it's the same thing. What you're going to look for is that whatever number you have is your anchor

07:45

go all the way across. You'll get all of these synonymous measurements, so yield rate again tends to be tends to be more complicated because you have to incorporate your rework back into it. So it's not only how maney defects did you observe, but how many defects did you fix that? Still went out

08:03

where his defects per 1,000,000 opportunity is, How many chances were there for a defect and how many defects were there, actually,

08:09

So with that we have a homework assignment. You're gonna order food. Um, I don't care what That's not part of the assignment. You're gonna order food and then what you're going to do while you are calculating or DPM Oh, for your food delivery is you're going to determine how many different opportunities you have.

08:26

So if you order your food online, how many different fields do you need to fill out to be accurate

08:33

to get your order if you call in and how many different questions do they ask you? I want you to identify any defects in your experience, including your quoted delivery time, because if they say they're going to be there by 5 35 they're there at 5 36 this is a defect,

08:52

and I want you to calculate the DPM. Oh, from that. So go through the entire exercise. Identify defects,

08:58

Bonus rounds if you identify polka, yoga or fields that you have to fill out. But really, the take away here is what is the DPM? Oh, for your favorite delivery place.

09:09

Today we went over why it's called Six Sigma. We reviewed calculating DPM. Oh, and we reviewed using a Sigma table and in our next video, we're going to review the lean principles, so I will see you guys there.

Up Next

Similar Content