Ito University — Opening Remarks, Cromwell, Connecticut

This is a very special moment for UTC, and I really mean that quite sincerely. I'm delighted to be here with you this morning and also for virtually the entire week to share the experience with you and to learn as well. It is a special moment because for the first time in many, many years, and it is at least a half a dozen, we are going after UTC-wide education in a formal and focused manner. We really haven't had any since the early 1990s. It is an even more special moment because we begin this effort with Ito-san and with Ito University, and we signal in doing so without qualification our commitment to quality and to reliability of performance of our products and service as our first and most important priority as a company.

I might note just in passing, as we start here, that this grand individual, Ito-san, our great friend, moved here to Farmington, Connecticut, with his wife then of 39 years, now of 43 years, four years ago, July 1994. That was an extraordinary move, I think, from anybody's viewpoint in that Ito-san was at that time 66 years of age. He is now 70 years of age. He had already finished a 34-year distinguished career at Matsushita Electric as its corporate quality assurance executive. And I don't think so many of us, at age 66, would pick up and move halfway around the world into a foreign culture with a foreign language. And that's why we owe so much to this man and to his wife, and we are thrilled and happy and honored that he and Mrs. Ito live and reside here in Farmington, Connecticut.

I'm going to talk to you today about three ways that our world is different now, compared with only a decade ago. I think these are powerful and accurate and important changes. I am also going to follow those three ways the world is different with five specifics, some of which we will learn about this week.

The most significant way our world is different is what I call a process change revolution that has been developing since the late 1980s. And to take a very sweeping perspective for just a moment, I think the history of the post-war period and the world economy has been American research and development, but focused by big American companies and by our government virtually entirely on product, to the exclusion of process. In 50 years we invented practically everything. Each one of us here is old enough to know what life was like in the 1950s and 1960s and 1970s. And even the 10-second memory of your childhood will show you the impact of American R&D on the world economy in the last 50 years.

But what we didn't do ever is focus on process. And the Japanese did. They often took our own products and innovations in technologies. They re-engineered the process of those products and technologies and then, aided in part by the strength of the Reagan dollar in 1985-86, they invaded our markets aggressively in selective industries: semiconductors and automobiles, to name two.

We kept our markets open, we did the American thing, and we adopted the changes that we had to no matter how painful those changes were at the time. We took the best practices from others. We incorporated them in the American experience. We learned and responded and we have built the most powerful economic recovery, I think, ever in the history of the planet.

So the last dozen years have been a history of process change revolution in the American economy. It is also true that the pace of change required to compete has consequently increased dramatically. I think in my own career, five, 10, 15, 20 years ago, how fast things happened in terms of process change, product change. It has gone up by factors of 10, not in terms of the magnitude of change, but in terms of the rate of change.

So one way our world is different is this process change revolution.

The second way our world is different is that unlike all other management methodologies, this series of disciplines that I collectively call process change, which includes kaizen, quality, lead-time compression and many others, has a rock-hard foundation, a theory, simplicity, practice and implemented delivery. I think about UTC education I have had in the past, and it was stuff like the One Minute Manager. We have all read Peter Drucker with his accurate, but nonetheless sweeping, generalizations. We all know those wonderful consultants' four-square matrices, you know: high/high, low/low, low/high, and what box you are in is what you do. Life is real simple and real easy. And why this set of disciplines I call process change all together is different is because it is instead discipline, specifics and everybody in the entire organization, whether it is a factory or UTC, acting together with a common language and a common understanding.

I want to tell you a story, because I have had quality education for a long time, as have many of you. My first exposure to quality education was back in the mid 1980s at Otis Elevator Company with Phil Crosby and his four absolutes. And I will bet that each one of you could pass a test of reciting those four absolutes. Of course the first is that the definition of quality is conformance to requirements. The second is that we achieve quality by prevention, not by correction. The third is the standard is zero defects. And the fourth is that the cost of quality is the cost of non-conformance.

I hated Crosby. I couldn't stand it. I was frustrated because I didn't know what to do. And I finally decided that what was wrong with Crosby was that it was theology, which is to say it was belief absent information or hard fact. It was cheerleading. It was theology. It was belief. It gave us the goals, but come the end of the day, the end of the course, nobody had the slightest idea of what to do next or ever except the priest. There may be some Crosby likers in this group, and maybe my words are too harsh here, but I want to make a clear distinction between what we had before, which really was belief absent course, and what we have today, which is precision and implementability. And I really believe that that is the most important part of what we will learn about this week.

The third change in our world today is that we don't have any choice any more about whether or not to implement these process change disciplines. Everybody likes to take credit today for the American economy, and I am going to make this little sideline for a political comment for a moment. The whole world says the American economy is great and they all say that they did it. It starts with the president, and Mr. Greenspan for sure, and the American Congress, and many others. But I know where the credit actually resides. It resides first in this process change revolution where we are taking 50 years of Japanese practice and we have jammed it into the American economy in a single decade period. It is the intellectual discipline that we have taken from others.

Security analysts have forced this kind of implementation of these process changes for us. That is the third point, which is the presence and the practice of security analysts. I talk to them all the time, and is it is very tough for us as competitors or peers to get the jump on any other company. And this is the change now versus only 10 years ago – the presence and activity of these security analysts.

Every time we get a new idea, Mr. Welch has six sigma or anything else, we brag about it. That is a learned survival technique with the analysts. You talk to them. You always have to have a new idea. So you talk to your divisions and say, "Well what's new?" And you find some new technique and we go and we talk about it. The problem is the next day they are in everybody else's face as well, saying what are they doing about X and why are their results not where they should be? And so what has happened is the presence of these analysts has forced very aggressive promulgation across corporate America of new ideas and new innovations.

So to summarize, with process change and this revolution, we have innovations that work, and with security analysts, we have the means to promulgate these disciplines and to force change across the entire economy, the whole society. And that is why we have the economy we have today. Not because of Mr. Greenspan, not because of Mr. Clinton, not because of anybody else.

It is a good place right here, I believe, to look at our own score card about changes and forcing changes and improvements. Our stock has gone up faster than many other companies'. So that is a positive. But the fact is that improvement is what is driving assessments by others at UTC – Improvement, not absolutes.

We can note that on key measures – and there is one I prefer the most, which is operating income to sales ratio for our company – we still rank after all these nice improvements and the stock price four times higher, we still rank number 10 out of 13 peer companies on that measure. We like to tout our improvement. I go see the analysts and say, well six went to nine percent of income to sales, and isn't that great? We tout it, and analysts and investors do cut us a break and say improvement is good. But the fact of the matter is we still rank only number seven out of the same 13 on the improvement. Number 10 out of 13 on absolute. Number seven out of 13 on improvement.

So we have a long way yet to go in bottom line results. We also know that bottom line results are only the score card at the end of the day after we have implemented the kinds of teachings that we are going to learn about this week. That is what actually makes things happen. Great products, great engineering, low costs, great manufacturing, effective marketing and presence in the marketplace. That is how we do what we do to get that result come the end of the day.

Now these five specifics I want to mention this morning, each one of them is in the process change category, but not all are about quality. In fact only three are, and therefore not all will be in our curriculum this week. The first one, which I think is not in the curriculum except by reference, is kaizen, or lean production, or what we have come to call at UTC, process re-engineering.

It began for us with Shingijutsu at Otis in Bloomington, Indiana, in 1991. Seems like a long time ago now. We moved to Pratt with Shingijutsu the following year, 1992. Sikorsky did agile manufacturing first in 1992. We had UTC's flex program in 1993. And together all of these things implemented fundamentally classic lean manufacturing techniques, and they are at the heart of most of the cost and profit improvement at UTC that Wall Street likes so much. That is the reason why the stock has done what it has done, in my opinion.

Specifically, these are amazing statistics. We are doing today in our US operations higher physical volumes of production with a work force that is one-third smaller. We were at 108,000 employees domestically in 1990, and we are today at 72,000 employees, a 33% reduction with higher physical volume. So that is one specific, and it continues, and we will never stop with process re-engineering. It goes on, but lots and lots and lots of gains have come from it already.

The second specific I want to talk about is very much at the heart of our quality efforts, and this one is what I call relentless root cause analysis. And I have a couple of stories here that I think are very persuasive. This is also definitely in this week's curriculum. The first story takes you back to 1986, and this is also where Ito University came from. This is right at the beginning. In that year we installed Otis Elevonic 401 elevators in a pair of Matsushita Electric's office buildings in Osaka, Japan. And for those of you not in the elevator business, Matsushita is our partner in Japan. We have 51% control, but without Matsushita Electric we have a long hard uphill climb in that most challenging market. So we put our new elevators in their building.

We were also only a couple of years at that point into the electronic control revolution at Otis. And the fact is that the elevators did not perform to Japanese market standards. We typically in Otis have higher callbacks, where a mechanic comes to work on the elevator to fix a problem, in the first year after handover, which is itself a problem. But even so, in this first year we had callback rates as high as 40 calls per elevator per annum. And that compared to a Japanese market standard of four-tenths of a call per elevator per annum. One hundred times out of sync with market standards.

Well, we did the usual thing, which was to dispatch our North American field engineers to Japan. Those men and women were then heroes in the Otis culture. They still are today. They are also, by the way, our comparable heroes in the cultures of each of the divisions of UTC – field engineers who fix things that are broken. And this is a complete aside: We want to ask ourselves a question and I invite you to put this in your head for a second. Ask yourself the question why is it that we have cultures that have high standards or high stature associated with things that break? Yet we did then in '86. We do now in 1998.

What happened next, after these good engineers got to Japan, was to me amazingly revealing because the Japanese company, Nippon Otis Elevator Company, took the field engineers, the heroes, the saviors, the people that fix them, they took them not to the motor room. They took them to the conference room. And they demanded root cause analysis right then and there even while elevators were shut down and while the place was literally on fire.

I will never forget that. And I learned right then and there that the American character, I believe, has in it the propensity to fix, to temporize, to get by. We do it with a smile. With the best of intentions. We are not bad people. We just tend to kind of make things work. It is Yankee ingenuity in the latter years of the 20th century. The Japanese character instead is to search for bedrock. And those cultures came together in 1986 in that pair of buildings, and the Japanese won, and it was correct that they won.

We got the elevators fixed, both in that building and in another pair of buildings, and also for the Elevonic 401 population worldwide, but not without some bumps and delays. They were bigger than we would have liked in hindsight, but we did it. And we did it eventually and in what I would come to call the Japanese way.

One of the bumps was a letter I got from Matsushita Electric's then President, Tanii, in 1990, eight years ago. He said, and I quote, "It has become clear that the inferior quality of your elevators is about to affect the quality image of not only Nippon Otis, but Matsushita products other than elevators. This situation cannot be left unattended."

For those of you not Japanese specialists, you should know that we trademark our product in Japan as National Otis, and National is the Panasonic domestic trademark in Japan. And so it is kind of like somebody else using the Pratt logo, for example, with the Sikorsky flying wing. We use their trademark. And he said, "This situation cannot be left unattended." And the year before that our good friend Ito-san, himself accompanied by Takahata-san, a member of the board of Matsushita Electric, had come to visit me in Farmington with essentially the same message. So that is one example of root cause.

Here's one more. We just dodged an in-flight shutdown on a Pratt 4084- powered 777 for Egypt Air toward the end of last year. I say we just dodged it because we were fortunate the flight was only Rome to Cairo. The airplane and the engineer had probably another half hour or so before we had oil pressure loss and engine over-temp. We lost 13 out of 16 quarts in the engine. And we lost it because of failure of a 30-cent part, a rubber O-ring.

The reason I tell you this story is the significance of what came after that. A few years ago and before Ito-san, I am convinced, that O-ring would have found its way to the trash. We would have had a new one installed, service resumed, Yankee ingenuity, no problem, away we go. And the problem might have repeated. Instead, in this case late last year, with Ito-san's influence, we got the offending ring back, and with some organized study we went across all of UTC.

We found out that we use 1.5 million O-rings a year. That is a lot of O-rings. We also found that in 1997 we had service failures of products and service 72 times last year across UTC, consequent on these O-ring failures. About half of them were mis-installation, that is, you put the O-ring in place and in the course of doing so you might slit it with your thumbnail or torque it or twist it and eventually it would fracture and break and you would have lubricant loss, and whatever other consequence. So half was mis-installation.

Another quarter was wrong part number. These things do look alike. Another quarter everything else. The particular Egypt Air failure was both mis-installation and inadequate supplier qualification of the ring in the first place. What to me is amazing is how you actually fix products with this kind of relentless root cause and Ito-san's idea that a failed component is a treasure. It is a gem that will yield its secret.

The third specific, and this is one we are also going to learn about this week, is quality assurance through control of individual processes. And I will make a distinction from the more usual American end of line accept/reject testing that we typically have associated with the inevitable mod shop for rework. And while I don't like to admit it, this is in fact our old friend Crosby's absolute – that quality is prevention, not correction. That is what this is. We do end of line accept/reject. The alternate is process control up front. The precise determination of the performance parameter of each individual process in making a part or component, and then the operator's self-involvement in inspecting and correcting, is what Ito-san calls QCPC, quality control process charting. I think you will be impressed by drops in QCPC in turnback rates. We will hear this week from 2000% to 50% over periods so short as six to 12 months.

The fourth specific, and again this is a discipline we are going to learn about this week, is the visual factory, what Ito-san calls 5-S. Those five wonderful words of sort, straighten, standardize, sustain, and – I love this bit – shine. I may be corrected from making those two terms, visual factory and 5-S interchangeable, because this is all new to me as it is to many of you. They mean a work place so tightly organized and so well laid out, where the performance parameters are so well understood, and the processes are so well signed, that the operation is immediately and transparently clear to any informed observer.

Now I am for sure not a factory expert. But seeing, as I do, many UTC plants, I know there is more of this in our future than we have today. One of the first things I look for is the visual factory, signage, and I quiz the operators, about the signage, what's going on, what's the process, what's the measurement, what's the accept/reject at the process level. We want the visual factories and those employees totally involved in understanding the performance parameters of the components they make. Our best factories do this all the time, as a matter of routine and regular course, but we've got a lot of factories, and we have a long, long way to go with QCPC and rolling back up the quality process to the individual process by the individual operator.

My last specific of five is not on our curriculum this week, but it is important enough for our future that I want to take a moment or two and mention it. This is lead-time compression and supplier partners. And a lot of this is going to come in the next year or two, and to me it is the next really big initiative in UTC's efforts to improve and make our company stronger and better.

I will do this again with a story. It is about Kent Brittan, known to many of you personally, and this time about Kent's visit. He has been at his job now for about a year and a half as UTC's purchasing and supplier relations executive. About a year ago he went to Honda Marysville to learn benchmarking. He came back and talked to me about a $10 billion asset building 400,000 cars a year. And Kent's eyes were about that big around, and we searched together for a few minutes about how to describe the experience that he had had. And we finally produced a word, which was revelation. It really was. Four hours in the front end, four hours in the back, 16 hour cycle time to build the car in the middle, with the result that from first cost to revenue the cycle was 24 hours, which means that inventory turns daily, every day, or stated ultimately, 365 times a year. And chief financial officers present, there's a couple of you in the room, might want to take notice about the new target of 365 turns per year.

But more significant, much more significant than that fact of 365, is the fact that Honda makes in the car only the motor. Amazingly, only 200 suppliers make everything else that feeds the front end of this four hour window to a $10 billion asset. And why this was for me and for Kent a revelation was that those statistics put whole new levels of definition on what we mean by just in time and zero defects. You can't have a failure when there is that much critical dependence on an integrated supplier relationship.

Now by comparison – I don't mean to ever do more than illustrate with our best friends here, never criticize or anything like that – but by comparison Carrier Collierville, which by the way is a great plant, turns out a million residential condensers to stuff in your backyard a year. High throughput, great quality, great operation. But they have a 12-day manufacturing cycle time, over the course of which we expend characteristically four factory direct labor hours. Compare that to the 16-hour cycle time for Honda Marysville, and we are making the condensing unit in the back yard of our houses.

Now give Carrier credit, lots of credit. There is a precise and disciplined plan to cut these lead and cycle times across all of Carrier, and they have a committed program to take half a billion dollars out of Carrier Corporation inventory for about one half of the present day total, all of this by virtue of supplier integration, cycle time compression, lead-time reduction.

So those are five specifics. We will hear about three this week in detail, but I wanted to put kaizen in the front for introduction and put supplier partnering and lead-time compression at the back end because they are so important to our future.

Now what does all this stuff mean for all of us? First, there is the umbrella overall of process change. Process change, I think, from the point of view of any big American company, is a survival issue. We change, we fix it or we die. There is unambiguously no doubt about that whatsoever. Also, process change is what has given us the gains we have experienced to date that we like so much. It is also what gives us confidence that we have much, much more process change in our future. But that is overall, that is everything.

What we are here to talk about this week is quality, and quality as well has a big financial benefit. We talked about it specifically, at least officially, in the accounting records as warranty costs. We spend at UTC about a half a billion dollars a year on warranty costs. It's a big number. Operating income by comparison is almost two and a quarter billion. This is 500 million. A half a billion versus almost two and a quarter. It is the cost of warranty that we actually disclose, account for, keep track of, which way, way understates the costs of non-quality. We don't include scrap and re-work in that number. We don't include customer dissatisfaction in that number. We don't include market share impact and many others as well.

Put it all together and I am convinced the cost opportunity in front of us of real quality is easily $2 billion a year. We could double operating income in UTC with perfect quality. That's the challenge I put before you today, both this room and across all UTC. There are also some charts around the room here that had that $2 billion number on it to talk about the above the line warranty costs and the below the line indirect impact like scrap, re-work, customer dissatisfaction. But I might note, having made an overall statement and having some charts around the room, we also know that no program ever works without precise measurement and clear goals and so there is a long translation process between the overall $2 billion and what to do tomorrow.

Now, more broadly, what's in front of us at UTC today as a company, I think, is an amazingly bright future. It is incredible how strong the future is in front of this great enterprise. Let's look at today for a moment. We start today number 10 out of 13 in profitability. We have products – in fact, 40 – performing in service in many, many parts of our company. We know that. We accept that. We also combine these with the strength of our market position and our global franchises, which are as strong as for any single company in the entire world. This is an amazing enterprise.

We also have a knowledge of and belief in process change. We've been at this now for seven or eight years. We've done the kaizen process re-engineering thing, not entirely, but substantially. Ito-san has lived with us here in Farmington for four years. Including his time with us in Japan, he's been with us for nearly 10 years. We know about this. These techniques are real, they work, we believe them, and we have to simply implement them across all of our company.

Sometimes I'm reminded of my own time at UTC, as all of us are. We all learn from our experiences. I sometimes say half in jest that good careers are starting out with great market franchises and poorly performing assets. That's the best possible combination to be, young division executives. You love the big strong market performer, but sick division. That's exactly what UTC has been for the last decade. It's been the sick division and it's come a long, long way from its low points with restructuring in 1991. But it's got phenomenal amounts left to go in our future.

Let me close with one tough topic for a moment and offer some advice to this audience and others. It's on a very specific issue in our company. We had our executive conference in Hawaii last October. Many of you here were there for that meeting. We did the usual conference survey of what's best, what's worst about our company, all the usual kinds of questions. The interesting fact that the number one needed change in our company in this survey of participants was to strengthen management.

We combined that experience with a comment I had in a long session with an experienced and senior executive over the last couple of weeks, and he said to me, "Look, the vision thing from you is fine. We're very happy with all that. But without more specific instructions from me personally about what to do, we were in jeopardy as a company of acting incorrectly, because of misinterpretations of the vision thing from me by managers, or because details were not clear enough.

In other words, great visions but not enough precision. And my advice to you is that both of these experiences – the need to strengthen management and the need for more instructions – are in the category of upward delegation. In other words, land the problem in the boss' office, wait for further instructions. That's been UTC's past. It is not UTC's situation today. And once and for all and for this category of topics only, my door is closed. It's always open, but not on this one, because there is in front of all of us a jointly held vision about improvement.

What we need to do now is specifics, and the practical matter is that I'm not going to, I'm just not able to, devise thousands and thousands of detailed plans across all of UTC. That's not my job. That is your job as individuals and executives. What we need in our future is not upward delegation. We need people that work within broad policy and broad guidance, and to harness the resources and creativity of all 200,000 of our employees worldwide. We do this already in the best of our companies. We don't have upward delegation. We have policy, practice, strategy, belief and vision we all agree about. And then we have people that are off at the edge of the empire with their own initiative, doing the correct and proper thing, and not turning back and saying, "Strengthen management," not turning back and saying, "We need more instruction," but instead, are acting with personal initiative.

We have a great situation. The stock price went from a split-adjusted low 20s half a dozen years ago to nearly a hundred now. I don't ever give stock tips. That would be wrong. But I'm not selling stock at all. This thing is going to go to 250, and it's going to do it within the reasonably foreseeable future of the executives and participants and people in this room. We have such a wonderful situation of an amazingly powerful business base with defined methodologies to make it better. We know how to do it, we believe it, and now we have to go out and do it. And that's why I say buckle up and let's have a lot of fun together. So that's enough for cheerleading. Now it's time to go to work.

Thank you very much.