Hi, guys. Welcome back. I'm Katherine McKeever, and this is your lean six sigma green belt. Today we're gonna go over the Peredo Principles and Paraded Day. So we just got finished discussing hissed a grams, which is a form of graphical analysis specifically for continuous data.
Now, we're going to be looking at our Peredo analysis, which is a form of a history Graham for discrete data.
So if Peredo sounds familiar to you, which it hopefully does we discussed this quite obey in death in a yellow belt. Eso if you need to go take a peek back at those modules for a quick refresher.
So this is a Peredo diagram and the really important thing to notice. As you are looking at this, there are a couple of things that should jump to your mind. One is this is, in fact, a version of a history. Graham, this is what we call assorted hissed a gram.
So if you'll notice on the X axes,
our numbers are no longer presented in the form of a spectrum or consecutive data. They are now in categorical. I created two inch chunks to look at our U. S. Presidents Heights because I did promise you earlier in the course. This is the only data that we were going to be looking at was the U. S. Presidents height. Well, the U. S presidential data
the so a couple of other things to call out specifically as you're looking at your Peredo diagram. So one this is a sorted hissed a gram. So using that mode, we talked about where the frequency of, um
count of those different measurements. You'll see that the left side is going to have the most frequently occurring measurement on the right side will have the least Frequently. The second part is, is this is a three axis diagram. So if you look at your X axes, you see the hype and inches.
If you look at your why axes, you see the count of presidents,
and if you look at your y two, you're going to see the cumulative percentage, and this is going to show us how the numbers add up. So if you remember from our yellow belt, the Peredo principle was the idea that 20% of the instances
consume 80% of the resource is so, for example, 20% of the complaints take 80% of the
customer service resource is or 20% of your employees take 80% of your management time.
This has been demonstrated over the last couple of 100 years, both economically and statistically. So that's why Wilfredo Peredo got the Peredo principal named after himself
checking out an analysis. This is a graphical analysis for our discrete data. It is a form of a hist A gram. Um, how we get here, given that it is a form of a history, am going back to excel. We're going to go to our data analysis tab we're going to select hissed a gram.
The where it gets a little bit different in Tricky is of course, you're gonna pick your input range. This is going to be your data. When you go to your output options,
you're going to want to select Peredo and cumulative percentage as well as chart out book output. So all three boxes. So, Peredo, that's going to give you that sort of hissed a gram where the most frequently occurring numbers are on the left
cumulative percentage. That's going to give you that second. Why excess which will tell you where your numbers stack up.
And of course, char output means that Excel draws your chart for you when we get there. So now you have a Peredo analysis. Don't get excited. This is the same one we just looked at. How you're going to read this? The first thing that I have to do is apologize to you because I promised you early in the course that we're only gonna use the U. S. Presidential data.
So in doing so, I did not change any of the data to make this easy to read.
What that means is this particular example is not the most clear and easy to read that you could look at. So we don't see a really strong manifestation of the Peredo principle here where we would expect to see a much higher proportion
in the first couple of bars.
And then what you're going to look for in your Peredo analysis is either your 80% where 20% of your activities consume 80% of your resource or a break. So when I say break when you're looking at this specific Peredo analysis, it has a nice, smooth
rainbow shape to it. You have kind of this gradual
rolling hills If you're from the Midwest curve to it, what you're going to see in more traditional Peredo manifestations is is a much steeper initial couple of data points because we do expect the first
20% of bars or categories toe have the bulk of the data points, so a much steeper curve and then a plateau. So if you're from the West like Arizona steep sides plateau to top that joint where it goes from, Ah, very high acceleration to a more plaid toad.
Cumulative percentage is called your break.
That is where you want to draw your line in the absence of
and the absence of a clear break. For some reason, the other way that you will use this for determining where you're going to invest your resource is because this is part of your prioritization and your analysis for what is causing the majority of our pain points
in our process. You can also look at 80% so look at where your 80% is. Draw a line straight down,
and we're going to say that 80% of our U. S presidents are within 68 74 inches. If you look over on the right, you can see we've got 64 66 greater than 74 so that gives us a range of 80% 68 74 inches six inch range.
You can use that as well. If you were to look at this from a categorical perspective, so let's say instead of 72 this was complaints about color. Hopefully, this is ringing a bell from the examples we used in Yellow Belt. That is where you would want to invest your lean six Sigma
effort because that is where you're going to get the most bang for your buck.
So if I were going to make presidential pants, if I were going to be the most efficient with my resource is, I would make pants with an inseam for a 72 inch tall gentleman. That's where I would get the most return on my investment. So with that, of course, you have a homework assignment.
Um, I want you to create a parade, a graph, so
hopefully you have the data analysis to a pack from our previous lectures a couple of lectures ago, and that means that you're going to use the hissed a gram option. In order to do this, you do need to have discrete data, and my recommendation is customer satisfaction scores or complaint categories.
I love using complaint data because you can very, very clearly see
the majority of your clients complain about the same minority topics. So my pizza was cold. My order was wrong. The color was wrong, those sorts of things. And ideally, for a good Peredo analysis, you're going to need to have more than 30 data points.
Um, and that's just because that's where you get a nice clean break
If you don't get a clean break, or if you are not seeing that 80 20 rule, you can always expand your data points that then I would ask you whether or not you've done something, whether or not there is something funny and how you do your data sampling to look at that.
So with that today we went over performing a Peredo analysis. Do you know how to get the graph to spit out of Excel
in like 10 minutes and you know how to read it? So you're looking for the 80% in your cumulative percentage or your break eso that where you go from a rolling hill instead, it you're going to have a cliff and a plateau. That's what you're looking for
in our next module. We're going to go over sampling techniques.
So now we're going to start looking at How do we pull out our samples to make sure that what we're analyzing, in fact reflects our population, So I will see you guys there.