You will recall a recent column where I addressed the question of “Why hasn’t GM learned from NUMMI?” I received a lot of responses to that discussion which convinced me that many of you would like to hear more about it. Rather than just give you more of my thoughts, I decided it would be a good idea to get a GM insiders’ perspective on the matter. So, I asked a friend of mine from GM who is very knowledgeable about the company’s efforts to make the most of its experience with the NUMMI joint venture to give the question some thought and provide his prospective. Lou Farinola recently retired from GM after 39 short years with the company. From 2002 thru 2007 he was responsible for leading GM’s Global Manufacturing System. Lou was also at Cadillac in 1986 when the folks I mentioned in my column returned from NUMMI and first attempted to install TPS at GM. So, Lou is exactly the right person to share his knowledge and wisdom about what happened, what didn’t, and why. Lou and I hope you find the following exchange of value.
So, Lou, how would you answer the question, “Why hasn’t GM learned from NUMMI?”
Lou: Well, John, as you stated in your December blog, a simple answer could be that GM has in fact learned a great deal from NUMMI and continues to do so today. It is also true, however, that it took us a long time to really get it and to fundamentally change the way we designed and built vehicles as a result.
John: Give us an idea of what happened on the GM side in those early days.
Lou: When the folks you mentioned who had been stationed at NUMMI to learn for a couple of years came back to “mainstream GM” in the late 1980’s The Machine That Changed the World book had yet to be published and there were still very few people who recognized that Toyota was changing manufacturing forever. I worked for the Divisional Personnel Director at the time and I recall the conversations about what to do with this group of executives that had just spent two years learning TPS. Some felt that they should be consultants to our manufacturing team and others, my boss included, believed that they should be allowed to run a plant and demonstrate the new system. In the end it was agreed to use them as consultants. This might have worked in a receptive environment but unfortunately they found themselves in an organization that saw very little need to change.
John: I remember also hearing that they requested to be placed together, so they could support each other, but ended up being dispersed to different areas of the vast GM system. So in the beginning there was just too much resistance to change for them to overcome?
Lou: Yes, they were together for a short while but then separated to different areas and resistance to change was a major factor, but I would also say that the team tried to take the GM organization to where they thought it should be with not enough regard for where the organization really was at the time. For instance, they put a lot of emphasis on the power of the team on the plant floor. Certainly, the team concept is a very key part of TPS. But the Cadillac assembly plant had just been built and was organized around teams and team coordinators so the leadership believed that they had already “done teams” with the result that the NUMMI team concept didn’t appear to be anything different from what we were already doing. Without a doubt, the organization’s understanding of teams, team leaders, and most importantly the management behaviors required to make it work did not mature until many years later. What’s more, there was no comprehension of all the other elements of lean and the integration of those elements into a total lean system.
John: So what did it take to start to change the company?
Lou: Actually it took several things and several years. Let’s start with GM Europe. In 1992 a new plant was constructed in Eisenach, Germany. This plant was led by a charismatic Lean Thinker who had been trained in the Suzuki Production System at the CAMI plant in Canada. The plant was built to be a showcase of lean and is still one of the best in GM today. GM Europe became convinced and the Eisenach plant became a training ground for GM manufacturing leaders throughout Europe and GME leadership amplified the effort by hiring numerous Toyota and NUMMI managers to create a critical mass of Lean Thinkers to make the change last. They even hired a leading TPS sensei from Toyota Japan to help train the leadership.
In the U.S. we started getting serious in 1994 when the same Cadillac assembly plant was selected as the “Lean Vision” plant and assigned to demonstrate that an existing operation could be made into a lean plant. I was the General Assembly Area Manager at the time and we had a Material Director who had spent two years at NUMMI. An important point to make here is that the people who were returning from NUMMI assignments were beginning to move around and up in the organization in line positions where they could really begin to make a difference. Anyway, this Material Director along with some central office help assigned by the Manufacturing Manager (the same guy who ran Eisenach and was now promoted back to the U.S.) decided to start with fundamentally changing the material management system for the plant. They worked on internal and external material flow, moving to much smaller lots and pulling rather than pushing material to the line. We set a target of no more than four hours of material at the operation, no more than 1 day of material in the plant and all material within the operator’s footprint. This turned out to be a great place to start because it demonstrated positive business results and also created a few more material jobs which most people considered desirable work. The improved material flow allowed me and my production team to embark on several other strategies like work station set up, line compression, fixed position stop and installing small buffers in the conveyor system. I could talk about what we did for hours but the point I want to make is that for us, the material system was the right place to start. I should also say that as managers returned from NUMMI and took assignments in other GM plants similar things took place at virtually every location.
John: So you had the U.S. and Europe going on parallel paths.
Lou: Yes, parallel but I would say not coordinated.
John: So how did the single Global Manufacturing System get started?
Lou: Good question and here is where leadership played the key role. In 1996 GM had decided to build four new assembly plants in four countries: Poland, China, Thailand, and Argentina. The CEO at the time, Jack Smith, was very committed to a strategy to make all GM’s operating systems globally common. He directed that since all these plants were going to be GM plants, he expected them all to have the same operating system with common principles and elements. Now by this time we had a good number of people with lean knowledge and we needed to get everyone on the same page. So representatives from North America, Latin America/Africa and the Middle East, Europe and Asia convened in Eisenach to work with the teams building the new plants to develop what is today GM’s Global Manufacturing System.
John: From all reports and according to any number of objective measures, GM manufacturing has come a long way. And it’s clear that everyone is speaking a common language as it relates to lean. How far along would you say GM is today on its lean journey?
Lou: Well, obviously we still have work to do, but I would say that there is a very good foundation in place and real progress has been going on for the last 10 years or so. Looking at each of the Five Principles that make up GM’s Global Manufacturing System, I would say that we are pretty solid in the areas of standardized work, built-in-quality, lean material flow and continuous improvement. This last one is critically important because it fosters a mentality that we are never done. The area where we have always had the biggest challenge, especially in the older plants, is people involvement. Don’t get me wrong, we have plants in every region that are outstanding examples of this principle, too. But I have to say that we have more work to do to get to the level that you talk about in Managing to Learn.
John: Well let’s leave it there for now and see what the readers think. If people found this useful would you be willing to answer a few more questions about the GM experience?
Thanks, Lou. I have more questions myself, but I’ll stop here and we’ll see what our readers come back with. Who knows, maybe Senator Corker from Tennessee will join in.
Lean Enterprise Institute, Inc.