Early learnings in installing Demming's 14 point system and improving quality at Donsco Inc.

Art Mann  We're a foundry machining operation. We have operations in Pennsylvania, Belleville, Pennsylvania, here in Wrightsville, along with our own pattern shop, ops and machine shop. And we have approximately 500 people. And we're probably the largest fully integrated cast parts producer because we produce all our own tooling, from foundry tooling to machine tooling. So we can make a part very quickly from conception to finished part, sometimes within two or three weeks. And we've been in business since 1906, and we've just celebrated our hundredth anniversary. And the fourth generation of the family has taken over. My oldest son, Arthur, is now president, and the second son that's five years younger is now head of basically vice president of operations. When I first started here back in 1965, after I get out of the Navy, I came back and was actually going to go back to graduate school and study computer science at Carnegie Mellon. But I end up getting fascinated with the whole foundry process. And from that point at that time, there were only about 50 people and there weren't any engineers in the operation. And so I vowed after some time here that the only way to survive long term was to become an industrial based, highly engineered casting producer and to provide a finished part to manufacturers. So we started out years ago to define ourselves as the become a premier producer of finished cast parts, a one stop shop for customers who wanted the ease of buying a cast machine part with single source responsibility a lot less hassle. And so in the pursuit most of the issues you have with castings, because it is process based, you have maybe 40 to 50 variables that have to be controlled to make a good casting. From sandcast, from the sand that you use, to the metal, to the temperature, to the chemistry, to equipment, to the controls. It's a very complex system. We were always constantly fighting defects. There might be 40 defects that can give you bad casting. So the first inkling I had that there was some knowledge out there that really would help was listening to W. Edward stemming on NBC television and his white paper about national paper. And he said, using his techniques, that you would be able to tell if there was a problem, whether it was the machine or the man. And I thought, well, now there's a revelation, because I had no way on my own of ever being able to discern whether a problem we were having was the machine of the man. And so from that, we as a company got into statistical techniques. And I went to one of his seminars, which down to Washington DC. And there were 600 people there. And I saw IBM plant managers walking up to this guy and getting his autograph like he was some kind of a rock star. So I came back home and tried it. He said, well, you need somebody with a master's degree in statistics. And I said, well, in my circle of friends I certainly didn't know anybody like that. So turns out I had a professor friend, Franken and Marshall and asked Stanley and in the political science department, as it turns out, they were full of statisticians. So with Stanley and some other we actually created our own curriculum for statistical analysis and began to train our own people here. And I have to tell you this. One of the things he said after teaching our maintenance crew, he said, you know, these maintenance guys are actually better students than my students at F. M. And I've always felt that it's a good indication that the complex trades of working in plants and mill rights that have to know electrical theory, they have to know mechanical theory. These are highly skilled, very intelligent people, and their jobs make very satisfying careers. Except it's not celebrated like it should be. Anyway, from that point we began to apply techniques and began to understand our processes and we started making rapid improvements and the next big learning experience and we adopted by the way, we adopted Deming's 14 points because Deming saw that statistics weren't enough unless you had a management organization that built trust and drove. He said you have to drive fear out of the organization. That means you have to have an organization that has mutual trust, which is very easy to destroy and very difficult to maintain, but it has to be there. And oddly enough, our first customer from Japan, Honda. In their little motto for their beliefs, they said they believed in mutual trust and cooperation was one of their key values. I'll tell you a little about Honda and I think when I tell the story you'll begin to understand why their cars are so good in making the part. There was an engineer who was in charge of starting up their plant down in Carolinas and Atsumi's Card said he was engineer and he would have a little string tie and he'd come reviewing the parts, reviewing our whole process, asking us and working with us to control every dimension on that print. And he would be here evenings, weekends. And the upshot of it was that we began that as we got into production, we would once in a while get rejects because something was wrong or the magnet wasn't magnetized. And I told my machine shop manager that there's going to be a day when he would be upset when there was one defect. He kind of laughed. I said, you will find that when you adopt this Japanese philosophy that you can make perfect parts. And it was about two years later, after a year and a half of shipping about 200,000 parts without a reject, that one was taken off the supervisor's desk. It was a sample and was packed and the magnet wasn't magnetized. So when they went to start the lawnmower, it wouldn't start and they had to disassemble it. My plant manager almost had a heart attack over it and he was so upset and he admitted that to me that he never thought that they would happen. But it did and he was livid people that they would have let that happen. So they taught us that that kind of quality control of pokeyoking that you can you design the process so that you cannot make any bad part. You continue the story with our association with Japanese customers. The next one we picked up was Toyota and this was a very complex power steering pump which was hydraulic, which required robotic assembly of the cores. It was a difficult part. They had spent years and a lot of experimenting to get this part just right in terms of metallurgy, in terms of quality. And when we initially sat down with our engineers and they had selected us as a vendor, we had a little bit of raging argument over whether or not our process of a Cupolo Melbourne with a large forehard would be robust enough process to have the tight control over the medallurgy that they thought was necessary. And after a day of wrangling about that yes we can. No you can't. They came down the next morning to breakfast and said how about if we buy you the electric furnaces that you need? Well, those furnaces were 210 ton holders and cost about a million bucks back then. And our answer was well, sure. And they said well, we will buy them for you. You can install them and you can use them for other customers. But we want our parts run through those furnaces and make part of the process. And that's what we did. And they would come in and monitor our production to the point where we finally got they insisted we cut 1000 parts in half and inspect the interior and if all thousand parts were perfect, then we would no longer have to inspect them, which we learned how to do. And that's what we did. So now we could make hydraulic parts without having to look. And we were shipping at some point 60,000 units a week. So with the Japanese, you always had a very intelligent, highly experienced engineer that you were working with that they shared what they learned with you. And they only asked that if you develop things that we would also share it with them, which we in fact did. They copied our Gating system after they found out that they used to supply us a pattern saying they had their process proven and they would supply them afterward. One set that they sent us, our scrap rates went up. We said we would like to just do it the way we would do it. And we did. And the scrap rates went down and the engineer said would you share that design with us? Because we think that would be useful. So it was a very symbiotic relationship with trust and if there was a problem, they would come up and they'd say they have the problem analyzed. You would deal with it. We would have a resolution, we would fix it, and off we go again.