Dr. Liker is a nationally recognized authority on lean manufacturing methods and Professor of Industrial and Operations Engineering at the University of Michigan. He is an expert on U.S. and Japanese differences in manufacturing and supply chain management, and co-founded the Japan Technology Management Program at UM. He had a discussion with me on PDCA in this related podcast and transcription: PDCA The Toyota Way.
Joe: What makes PDCA, something other than just another problem-solving methodology like DMAIC? Is it a cultural thing? Is that what it’s all about?
Jeff: Yes, I think it’s a cultural thing. I’ve worked with various trained Black Belts and learned a lot from their Six Sigma Training and the projects they did. But the way I would look at DMAIC is that it’s very scripted, mechanistic approach. The underlying assumption which I think is a strong Western, scientific assumption ?? is that if you can understand the phenomena well enough through a combination of data and analysis of that data, you ought to be able to predict what’s going to happen.
The better you analyze the data, and the more you are precisely trying to identify exactly what the cause is, the better able you are to predict what’s going to happen. If you get it right then you’ve solved the problem, and like I said, the case is closed. You’re like the electrician. You put the tools back in your toolkit, and the case is closed.
In the Toyota way of thinking, again, it comes from humbleness, and it also comes from a background of the company, which started in a rice?growing region of Japan. If you’re a rice farmer, no sane rice farmer would think that they could control everything. They can’t control light; they can’t control the weather; they can’t control soil conditions. So they’re constantly trying to struggle to adapt to things that they can’t control.
That was the environment in which Toyota grew. It kept that humble attitude that the world is just too complex, and we’re never going to be able to predict what’s going to happen. It’s kind of a false sense of security to spend an enormous amount of time collecting data and analyzing the data with ever more sophisticated methods. Because the precision is false precision anyway, and we don’t really know what’s going to happen.
If we can come up with a lot of different ideas and try those ideas quickly, and then, actually learn by doing and actually see what happens. Then we have to go the next step of actually checking what actually happened, and what actually happened isn’t just what happened the week after the Six Sigma project was done. It’s what happens over the next six months, over the year, and over more than that. Do we sustain the process? So you live with the process to understand it.
Then you now have to, somehow, capture that information in a way that it’s reusable, so that people won’t make the same mistakes and don’t have to start the learning process over again in three years, when the manager has moved on, and most of the people have moved on. They have no idea that you made this great intervention three years ago, and you’re aware that these 10 things that work and these five things that don’t work. That’s where you become a learning organization.
My complaint is not necessarily DMAIC itself, but rather, the underlying philosophy and the organizational approach, which is you send in a Black Belt, who’s like the expert electrician. They take the tools out of their toolkit; they fix your problem, and they may work with you to try to teach you a little bit while you’re there. Then they put away their tools, and they go away.
They assume that they if they did enough statistical analysis which often, in these Six Sigma projects that I’ve been involved with, and even when I’ve seen the Lean Six Sigma that comes out of Six Sigma, I sometimes use the analogy that it’s about 80 percent massaging data numbers and about 20 percent actually thinking deeply about the problem, and taking action, and learning from what happens.
It’s the reverse in the Toyota way. It’s about 80 percent thinking deeply, engaging the right people, trying, observing, figuring out what happened, and educating the people who are part of that problem-solving process. Then, there’s about 15?20 percent, which are the tools and techniques and the analysis. That’s generally considered fairly irrelevant. It’s not a big deal, the methods themselves. The measure of success is what a person has taken away from this process: the hourly workers and the supervisors. What did they learn from this?
Lean Sales and Marketing: Learn about using CAP-Do