DW: What were you manufacturing?
SW: Nuclear measuring devices for the oil industry. Essentially, it's a non-contact measuring system, kind of like a Geiger counter.
DW: So, you were back in the same boat
SW: Yeah, I spent all this money, all this time. All the things I knew how to do I had done. I couldn't rearrange everything again. I couldn't look at the software and come up with any new ideas. I had already employed the best consultants that I knew.
DW: Right So what did you do?
SW: I signed up for Porsche mechanic school in California. It must have been a weak moment in my life. I do amateur racing and there's a saying that goes: you didn't make any mistake when you spun the car and flew off the track; what you did was you went into the corner and ran out of talent. That's how I looked at it-I must not be cut out for this job, there must be something I'm missing. I couldn't figure it out
DW: How old were you?
SW: That was ten years ago; so, early 30s. Mechanic school wasn't a waste of time. I still use what I learned. I save 600 bucks doing my own tune-ups. But right before I left to go out there, someone said: "You know, in San Jose there's a software company that has been cre- ated to support the rules that are stated in The Goal, and by the way, the Goldratt Institute has just issued a self-learning kit that you might be interested in." So I went to my mechanic class, that was very fun. Then afterwards I stopped in San Jose, took a look at the software, and completed the workbook on the way home. I was so excited that on Monday morning I got my staff together and I said: "This is what we're going to do. We've got nothing to lose. It looks like it's possible. It almost looks too simple. Let's give it a try." They weren't very con- vinced. In fact they were pretty skeptical. I'd put them through a lot already. One more thing, huh?
DW: This was their first exposure to TOC?
SW: Yes. Short story is, it took us about a month to go through the training materials, which came with a tutor guide and a workbook for all the participants. I went through the tutor guide step by step, they went through the workbook, and eventually they said: "I think you're right, we can do this." So we started, and about two weeks later we began to see some things improve. Lead times were starting to come
362
down, our on-time deliveries were starting to go up. At first I thought it was just a fluke.
DW: What changed your mind?
SW: Well, a month later here comes one of my welders and he says: "Boss, I think my numbers are wrong. The lead time I've been mea- suring is now about a day and a half." I said: "How can that be?" We were still running more orders. I had even had to fire a guy in the meantime, so we were down resources. And we hadn't bought any new equipment. So I said, "Okay, fine, let me check and I'll let you know what I find out."
DW: What did you find when you examined the numbers?
SW: I told my welder: "You know what? You're right, the numbers are wrong. The lead time is less than a day." Same resources, 40% more orders, a fraction of the lead time. Took us two months to do that. Cost us $500. The company was a hundred years old and they had the best two quarters that they've ever had. One division that was losing a million dollars a month was now making a million dollars a month. If I hadn't seen it with my own eyes, I would never have believed it.
DW: What was the constraint you exploited to make such
a huge difference?
SW: We actually worked through about three of them. One of them had to do with the fact that we were sending everything out to put a protective coating on the pipes that held the measuring equipment. It was a step that had been added at some point by the marketing department, and it had developed into a constraint. So we had to go and find one or two more suppliers to handle the load.
DW: And there were others?
SW: One was the saws that cut the pipes. We offloaded some of the work to another machine that was just sitting there doing nothing. That saw ran at half the speed of the other saw, no one ever wanted to use it. But we identified just the right materials to run on it, which built just enough capacity to eliminate the saw as a constraint. And then the
363
paint department was next, we did a couple of things there. At which point the constraint shifted to engineering. We were waiting for some new products to come out, and that's kind of where it ended up.
DW: Do you believe that TOC is an infinite process? In other words, is there always going to be another constraint you can find and exploit?
SW: Theoretically, it can go on forever. But from what I've seen, it goes through one or two cycles within a facility, and then you've kind of broken the constraint in the production operation. Then it may move to, say, engineering. Then you can apply Critical Chain to the engineering group and eliminate that as a constraint, and then the next constraint usually is the market, and typically it's the existing market. Unless you're Coke or GE or whoever, you probably don't have a dominant position in your market. So you can still find room to grow. Finally, there are plenty of cases where, using the same capabilities that you generated using TOC, you can attack new markets that you never thought you could compete in. At that point, you're probably doing all you can handle anyway.
Or maybe it goes back to manufacturing again. Could be, yeah, and you definitely know how to deal with that by then.
DW: Alright So then you moved on?
SW: I actually went to Grant Thornton for two years and worked on developing other TOC skills and applying what I knew to an ERP [enterprise resource planning] implementation at a plant in Mexico, working with Navistar International. I did that for about two years. Traveled to Mexico a lot, gained about 40 pounds, got no exercise. But it was kind of fun. Then I went to work for a consulting firm. Within about a month I was put on my first project, involving TOC, at a manufacturing facility in Clarksville, Tennessee, where they made graphite electrodes for the steel industry. It was a big plant, had been there quite a while, and it was already their best plant of that kind in the world. They made it a challenge for us, saying, "If you can improve things here, then we'll consider applying your methods elsewhere."
DW: This was a large-scale implementation?
364
SW: Huge. The plant covered half of Tennessee, it seemed like, way out in the middle of nowhere. So we put a small team together. It was me and another guy and about half a dozen folks at the site, and we went through the exact same training I had done the first time at Ohmart/Vega. Was exactly the same concept, exactly the same ideas. The only thing different was the context. We had software systems we had to integrate-five different software systems that had the data in it we needed. We identified the constraint, and did all the usual things, like making sure there was a buffer in front of it, making sure the maintenance guys were giving it top priority so if there's any trouble they could fix things right away. We put a quality check in front of it so that we weren't wasting time processing any bad electrodes at that point in the process.