You Don't Need Formulas or Decision Trees

Improvement Insights Blog

You Don’t Need Formulas or Decision Trees

You don’t need to know Six Sigma Formulas or decision trees to make improvement. Here’s why:

“Hi, I’m Jay Arthur, author of “Agile Process Innovation; Hacking Lean and Six Sigma for Results.”

“Back when I got started in Quality (and maybe many of you), we had to learn formulas to calculate things for control charts and whatever. I spent five days in a control chart class learning how to calculate the formulas for every control chart manually. Now, that did not really teach me anything about the distributions underlying them, it taught me how to calculate them manually. What they didn’t tell us in that class is software can do that for you.

“Later on, I built some software to do that for me so I didn’t have to do all that any more, and i don’t have to spend a lot of time teaching people those formulas. Some people will argue with me that, “Well, I had to learn it that way; everybody else has to learn it that way.” No, that’s stupid. Software will do it for you. Formulas are built in. They’re universal: my software, other software… you’re going to get the same answer, right? You may spend a lot more if you buy somebody else’s software than if you buy mine, but that’s the way that goes.

“We also had to learn decision trees. Now, people do not think normally in decision trees; that is not how people think. If you’ve ever tried to teach people decision trees: “Well, is it a variable or attribute? Is it this or that or something else?” [They] wind their way down to choosing the right chart. Guess what? They have a hard time learning that and they get it wrong most of the time.

“That’s why back in 2006 the QI Macros [became] the only software where you just select the data, and when you select the data I can look at your data, run that decision tree through the software (not through your head, through the software) and let the software pick the right chart for you. This is a huge advantage when it comes to learning. To people learning how to do Quality Improvement, control charts, Pareto charts, whatever it is. Does this make sense?

“When we automate these things, this is what’s called “Low-Level AI,” right? This is what’s going to help us get everyone to a point where they can make improvements very quickly. I don’t believe everybody has to know everything to improve anything. If you’re like most people, you discover that over the course of time you use five, six, or seven tools all the time to solve most of your problems and you don’t need to know the other stuff. The other stuff was “Just In Case” training: JIC, “Just In Case” training, in case you might happen to be in a situation where that might be possible.

“Well guess what? If you get in that “Just In Case” situation, there’s lots of training online; you can go learn how to do it online, right? You don’t need to know it when you first start. I often find that that we train people in too much “Just In Case” stuff, and then they can’t figure out where to start sometimes. It’s embarrassing. It’s overproduction. We’re trying to put too much stuff in your brain and so you can’t learn as quickly as you would otherwise.

“With software like QI Macros that has all the formulas, it has the decision trees, the Data Mining Wizard will go out and do… I used to spend hours trying to teach people my process for using pivot tables to summarize data, and then control charts and Pareto charts and fishbone. I built that into the software; I automated all of that. Why would I do that? So that people can start making improvements. Learning how to do all this formula stuff and decision tree stuff and data analysis stuff takes time and can be confusing, but if I automate all of that for you, guess what? We can collapse learning to a day (maybe less) and get results at the same time.

“So just because we all learned it the slow way does not mean we have to teach it the slow way. Not necessary. I’m against it, right? I want us to make it possible for everybody on the planet to be able to do Quality Improvement like that, but they’re going to need some tools to do it: mine or somebody else’s. I just think mine are better because I’m always trying to find a way to collapse the learning curve and get people going in a big hurry.

“So that’s my Improvement Insight for this week: Let’s get rid of the unnecessary training that people don’t have to have, and let’s start using that time to make improvements, and once people get that that little jolt of electricity from making an improvement, guess what? They’re addicted. They’re hooked for life. That’s the goal. That’s what I want to do. I want to hook people for life on Quality Improvement because I think it’s fun and I think it’s a blast.

“Now, not everybody loves that. Innovators don’t love that, but improvers (which is about two-thirds of the population) they love it, right? If i can make it simple for them, they’ll learn how.

“So that’s my Improvement Insight for this week: Let’s create a hassle-free America. Let’s go out and improve something this week.”

This entry was posted by Jay Arthur in Improvement Insights and tagged , , . Bookmark the permalink.