# 3.3 Defects Per Million Opportunities

Video Activity
Join over 3 million cybersecurity professionals advancing their career
or

Time
9 hours 3 minutes
Difficulty
Beginner
CEU/CPE
9
Video Transcription
00:00
Hi, guys. Welcome to getting started defects per 1,000,000 opportunities. I'm Catherine MacGyver. And today we will review the operational use for DPM. Oh, how to calculate DPM. Oh, and I'll give you a quick bonus round on yield rate and how to calculate that.
00:20
So we in our last module, we went over what six Sigma means and what the signal levels mean from a statistical perspective, and that is all great and very sexy. And it looks great on your company website to say that you work at a five signal level, But what does that really mean
00:39
for you on a day to day level? If you
00:42
showed up to your office and your C O said were functioning the three signal level and we need to move to a four signal level. How would you know what that is or how much improvement you need to make?
00:56
That's why we use D P M. O, which is defects per 1,000,000 opportunities. It is an operational definition that reflects the number of defects. It is more helpful than signal level, especially when you're working with project teams or operation and process management,
01:15
because it gives
01:17
specificity
01:19
almost Everybody who has a lean six Sigma program knows six Sigma means 3.4 defects per 1,000,000 opportunities. You know that four Sigma means 6210 defects per 1,000,000 opportunities. But do you know what five signal level is? So this
01:37
that lack of awareness is the reason why we use DPM. Oh, it is easier to understand
01:46
before we capture how we calculate DPM. Oh,
01:51
the first thing that we need to understand is what is an opportunity? So when opportunity is a chance for a defect, there are a couple of different ways that you can describe this. It can be finished products. So what are the likelihood of your finished products getting out of the door where there can be some specificity
02:10
opportunities, air defined both by your customer
02:14
and your business? But for the example in front of us, our contact form? There are a couple of different ways that you can articulate an opportunity for this so you can have one opportunity, and that is where we have a contact form that is completely filled out. You could have
02:31
three opportunities where it's a little fuzzy, but there are three *** tricks for required fields
02:37
and then over and above is
02:39
bonus round, or you can have 12 opportunities. This is something that has to be defined by you as the process. My recommendation to you is that your opportunities never be less than your minimum data set. So, for example, looking at this contact form,
02:57
it could arguably want the one opportunity. But
03:00
I would advocate that this is no less than three opportunities because there are three required fields. When we're talking about DPM. Oh, and process improvement, the more specificity we can get, the better. So in this scenario, I would see that we should have
03:20
12 opportunities for defect,
03:22
and then we can later use where those defects are as an input into projects, election and projects for improvement later on.
03:32
All right, once you guys have determined how many opportunities are there to get too deep ham. Oh, you're going to capture the number of defects observed so on our contact form on the previous field. If we said that there were 12 opportunities for defects
03:52
and we got
03:53
10 fields filled out, we're then gonna have to defects over 12 opportunities to get to the DPM. Oh, you're then going to calculate that by one million? The reason why we use this is that means that we don't necessarily have to wait for a 1,000,000 opportunities to get our DPM. Oh, great. So
04:12
the example that I have is we have 4000 defects observed
04:16
over 15,000 opportunities. We calculate that by a 1,000,000 we have a deep, deep e m o of approximately 266,666. So that gives us a starting point. For now, we understand out of every single opportunity we have slightly over one and four
04:35
that are going to have a defect on them
04:39
as defined by the customer and the business.
04:43
When we're talking about DPM. Oh, it is my preferred way of reporting organizational accuracy, and the reason why is because it is very intuitive. You have 10 opportunities for defects, and you have nine accurate fields or one defect
05:00
that gives you a place that you know that you have a 10% defect rate,
05:04
and you can use that for your starting for your improvements. There is 1/3 calculation regarding your organization's accuracy rate and that is yield rate, so yield rate is the inverse of your DPM. Oh yield rate shows us the possibility or the probability
05:25
of having a defect free
05:27
finished product
05:29
to calculate yield rate. You're going to take the number of units times the number of passing units and then divide that by the total units with one minus the percentage of passing units times the percent reworked.
05:44
So the reason why we include re work in here is because we're talking about products that leave
05:49
your organization. These air finished products these are output. So even if it wasn't done correctly the first time, if we went back and worked on it again, then it is now a finished product that can leave the door.
06:03
This is the reason why I do not like using yield rate because it gives an artificially skewed,
06:11
an artificially high reflection of how your organization is doing so. For example,
06:18
if you have 10 units and 10 opportunities for defect on each of those units, so you have 100 opportunities for defect. If you have a robust auditing program or robust QC program, so quality control and you guys identify those defects before they leave
06:39
your organization. Before those outputs move on to their customers,
06:44
you are going to rework them and thus report a higher yield rate than if you reported your d P m O.
06:51
There are some conversion tables for yield rate to DPM. Oh, they don't necessarily include the reworked aspect. So with that,
07:00
I'd like you to take on a homework assignment for yourself. So the first step for this homework assignment is identify the different opportunities within your process, whether or not it's a contact form. And it has all of these field whether or not it's receiving information,
07:18
identify all of the places where you could have a defect
07:24
with those identify which of those opportunities would require re workers miss. So these were going to be crucial errors or inconvenient areas. So something that stops the output from being completed, or something where you can push the output through and it's just not optimal.
07:42
And then, if you want to go for extra credit,
07:46
calculate your beat your baseline DPM Oh, for this process. But really, the takeaway for this homework assignment is learning to identify opportunities for defect in your processes.
08:00
All right, So when we talk about DPM Oh, we're talking about an operational definition that is more effective to communicate defect rate, especially to people who have no six signal or statistical training. It can be used to convert into a signal score. That's gonna be our next module.
08:18
But in order to be effective,
08:20
your baseline numbers of opportunities need to be identified. You need to know what you're comparing against or the denominator in your equation.
Up Next