Hi, guys. Welcome to analyze base hissed A grams and defects prioritization. I'm Catherine MacGyver, and in this lesson you'll be able to understand the creation and application of a hist a gram and understand how to read a history gram for process improvement. So first starting what is a history? Gram, I'm gonna make this super easy for you. It's a bar chart
It's a bar. Chart stats. People like to make themselves feel cool and give them cool names. But never, ever let yourself feel intimidated when somebody says I need you to make a history gram of something, It's a bar chart.
and the technical definition is a graphical representation of the frequency of some count of data. So there are a couple of terms that we use in history grams that are a little bit different than bar chart and excel. We talk about bins and frequency, so bins give us
the range of data that's being displayed. The example that I gave you guys is actually going to be
nominal data because we're talking more in the categorical land at the yellow belt level. But when you think about continuous or discrete data. You're going tohave A a range of numbers. So the temperature was between 90 and 95 95.1 in 100. So that's going to be your bin.
is the count of items that fall within that Ben so of your data sets. If we were to say, do the 95 excuse me in the 92 95 the 95.1 in the 100 how many days did the top temperature hit? 92 95 6 Okay, 95.1 to 100
seven. So you're going tohave
two bars that have a six and a seven because you're only looking at the count of number of data points that landed within that
been. So when you're looking at, how do you determine what all goes in your bar chart?
You're history, Graham.
The sum of your data points has to equal your sample size. So if you have 20 measurements, you're going to want to have a cumulative 20 data points on your history, ma'am.
So, like I promised you guys were working and categorical nominal data here. We're gonna look at some sample data regarding customer complaints.
So with it being sample data, it's not necessarily ordered. So when you see hissed a grams later on, we start talking about them and more depth and green, though you're going to see them ranked a little bit differently. And that's where we're going to start getting some ideas about our distribution. So if this was continuous
data, we would say that this has a binomial distribution. Because you have two peaks,
you have your 87 in your 45. For the sake of this exercise, there is no associated distribution because we're working with categorical data. But so when you create your hissed a gram, you're going to want to have your Benz. So in this example, Ben Number one is
took too long to get to me. So
the customer was waiting too long and you had 36 complaints associated with took too long to get to me. So you're been took too long to get to me. 36 the number of complaints that met that. So that's how you feel your frequency and bends
again. When you're working with numerical data, you start seeing things like the bell shaped curve or the two humped bells of the binomial distribution again. But right now, since we're in categorical data, you don't need to look at the shape of the distribution in this
lesson. In our next lesson and Peredo charts,
you will want to and we'll talk. About what? The difference between a hist, a gram and a parade. Oh, is. But so now you've built your hissed a gram, you have your
bar chart. Um, Now you're gonna ask yourself, what would you do with it? So the reason why you build hissed A grams is so that you have a point of reference for defect prioritization
moving down the road. You use this as a poor man's way to check distribution. But for today we're gonna talk about it as if I only have the ability to work on two different process improvement efforts. Which ones would they be?
That's why you want to do a hist a gram. So remember when we talk about project prioritization, we want to score it associated with the organization's goals. When we talk about defect prioritization, we have started teasing out our X and R Y variables.
We have started. We have done our data collection. In this specific example, our data collection methodology
is going to be counts of. Customer complaints were going to use an affinity diagram for this one because we have the customer complaints to get down to those themes. And now, as you look at your customer complaints, if you had to pick two, you would wanna pick wrong product received. And color isn't uniform
to focus your process improvement efforts
because this will give you the largest bang for your buck when you're doing your project.
So with that, remember that a history Graham is not intimidating. We do it because it helps us choose where we want to focus our efforts. So it is a bar graph. It shows us a frequency. So we use this for defects. Um,
we don't necessarily were not able to use this for any of the other waste, but we definitely use this for defects.
Ah, and when we start getting to time measurements for anything that has a frequency associated with it, we'll use a hist a gram of their starting point and then remember in the future we start using this for some of the more statistical aspects, but right now, you don't necessarily need to worry about this shape of your history, Graham.
up is Peredo charts, which is a variant on hissed A. Graham, and I have found to be much more useful.