§1. Theory of constraints: when it works and when it doesn't [00:23]
So we were talking about theory of constraints. I gave the example of the Ford F-150 plant where they were trying to operate under theory of constraints — where you have some manufacturing bottleneck, some machine that slows things down. I asked the foreman where the bottleneck was, and he said, "Well, it's the people, where I put the people." It wasn't the machine, it was the people that were the bottleneck. Some people work faster than others.
When you have something that's a nearly balanced flow, it's silly to try to introduce a constraint so you can manage it. There's an important principle there: many of the management gurus and theories are trying to make it easier — to have a simple way so the manager doesn't have to think. Just add four to that time, plant manager, solve everything with theory of constraints. There is no one magic bullet. If there was one magic bullet, I'd come up with it and I'd be a rich man.
Why do managers get paid more than hourly workers? That was a big question in my mind when I first graduated and went to my first job at Bethlehem Steel Corporation. The managers had better benefits, better pay, better everything. I thought, why do they treat the hourly workers as a lower class of society? I did learn why: because you never leave your job as a manager. At Bethlehem Steel, we all had to punch in and punch out on a time clock. When my technicians punched the time clock, they went home and didn't think about anything else until they got back on the clock. As a manager or a supervisor, I was always burdened — no matter whether I went on vacation with my family, I was always still thinking about how I could help solve some of these problems back at work. Managers are paid to solve the big problems. You can't just give up all your responsibility and think you're going to get good pay for not thinking. But that's what a lot of these people think they can do.
There are times though, to get back to theory of constraints, when it works. I've got two examples. [Tom passes around a turbine vane.] This is a turbine vane I got at a Pratt & Whitney plant in New Haven, Connecticut. This plant makes vanes for the engines. Forty percent of Pratt & Whitney's profits come from twenty percent of their business of making vanes. The rest of it's making the whole engine, but the vanes are only twenty percent of the engine and they get forty percent of the profits from it. So there's a lot of technology in making that vane.
In this factory, you have the incoming castings, and they have to be machined. The tail end of that casting is called the Christmas tree. You can see on this side it has arms going out. That has to be machined very precisely. You cannot do that on any regular milling machine or lathe. You have to broach this. Anyone know what a broach is? You've seen a tap and die — it's a machined piece. We talked about reamers before; it's sort of like a reamer, only you push it through to make a very precise hole. You're actually machining by shaving the material. Your tool has a bunch of cutting edges on it, and when you push it through, it leaves a very precise hole or other shape. These things are spinning around at 20,000 RPM or whatever, depending on the size of the engine. If you get any vibration in this mechanical interlock, you'll get fretting wear, which causes the thing to eventually fail in fatigue. You have to make a mechanical joint that is near perfect.
The machine that would do the grinding or broaching of this precise Christmas-tree shape — let's just say it was a $200 million machine, and they only had one of them. They had all these other little areas where they were doing the pre-prep or post-prep machining to make that blade from a casting, and it always had to go to the center machine. They put the machine right in the center of the building and reorganized into cellular manufacturing. One of the constraints they had worked on was trying to make the parts take the shortest path through the factory — coming in the back door, going out the front door, they didn't have to crisscross in a zigzag pattern. They wanted cellular manufacturing, but it all surrounded this one machine that you couldn't afford a second one of. You had to manage the maintainability of that machine, because when that machine went down, you were out of production. That was one where you manage that big machine and the process will work better. You don't let that machine go down, because that is controlling your throughput. You can have a million cells out there, but you're only going to get the throughput through that one machine.
Another example: I had a Ford student at a Ford plant once, and he was looking at painting of the next generation of Ford Taurus bumper. It was going to be polypropylene oxide as the polymer — very difficult to get adhesion, and he was going to have to work that out. In that plant where they were making bumpers and painting them — automotive paint line, ultraviolet curing of the paint — the paint line was a $200 million investment. They only had one of them. You could have all the injection molders for bumpers, and you could have all the things that might process the plastic surface, but you still had to manage that paint line because that controlled the throughput. So these are two examples where theory of constraints works. But there are many other examples like the F-150 line where theory of constraints doesn't work.
§2. Six Sigma, certified Jonahs, and the buzzword problem [07:42]
People come up with these ideas, some of which have good scientific basis, such as Six Sigma, which is a form of statistical process control. It turns out people looking for work will create jobs for themselves where they teach Six Sigma. I get business cards from people when I go into industry, and sometimes on their business card they have this little symbol saying they're a Six Sigma black belt. It means they took a course — because none of the universities were teaching it, someone would teach a continuing education course in statistical process control, and they passed the quiz and now they're a black belt in doing statistics for manufacturing.
It's actually something of significance. It means they passed a particular skill set. Here's the certificate they get, and you go into someone's office and many times that plaque is on the wall and they're very proud of what they've learned to do. That's justified. Goldratt in The Goal, which is theory of constraints, doesn't do it exactly the same, but if you remember the picture of Herbie — Herbie was the story, and Jonah was the consultant to the man who was trying to run the factory. Goldratt turned the whole thing into a business of teaching people. He has institutes around the world where you can go, pay a hefty tuition, and they will teach you to be — you'll get a certificate saying you're a certified Jonah. You've seen these, right? Anybody seen certified Jonahs or certified black belts? Well, you will.
I don't know what a certified Jonah knows. If you ask me what skill set they learned — they learned that someone can take your tuition, talk to you for a few hours, and then pass you on without any increase in your knowledge. Hopefully MIT's not there. I don't have a whole lot of respect for the idea of a certified Jonah. They're going to go and find the constraint and manage it.
§3. Just-in-time and its limits [10:11]
Another tool of TQM is just-in-time. This is part of the Toyota production system. Originally, you're reducing your inventory and reducing your floor space by pushing your inventory carrying costs back to your supplier. You say, "I don't want to receive it until I need it." In Nagoya, Japan, this was a very important thing because Toyota didn't have a lot of space in some of their cities. Toyota's suppliers, many of whom are affiliated with Toyota, would have to deliver every hour or every two hours. They would not have a production quantity of more than one or two hours of stock.
How would you like that if you were at Boeing? You can't build an aircraft with half a million parts with only an hour or two of stock. It doesn't work exactly for every industry, but it does reduce floor space, it shifts your inventory costs, and there are all kinds of stories of when they have a snowstorm in Nagoya, which doesn't occur very often. I got caught in a three- or four-inch snowstorm in Tokyo once in 1980 or so. I was trying to get to the airport to go home, and it's like an hour-and-a-half bus ride from the bus center. I got to the bus center, and because it had started to snow, they shut down the bus center. So I'm an hour and a half by bus from the airport, and I wanted to go home. I found a taxi driver who was willing to take me. It took us six hours on the road to get what the bus would usually take an hour and a half. It was the most expensive taxi ride I've ever had, even now. But I got there on time because the planes were all late too.
Just-in-time — there's a General Motors story, west of Cleveland: an automotive assembly plant where they were going to build Chevy Cavaliers. A new manager came in and said, "We're going to do just-in-time. We're not going to have more than one hour of stock for any part that goes on the automobile." A couple of people designing this assembly line said, "Well, we have some bolts —" and they could keep a week's supply in a box no bigger than this. "Nope, we're only going to have one hour." So you're going to have a box the size of a pill box — you know, for your prescription pills. That was what that manager said, and he was going to enforce it, even if it made no sense, because he was using just-in-time, and at the time it was one of the latest and greatest ideas. The point being: you have to use some common sense in applying these things. That's what managers get paid for.
§4. Agile manufacturing, the Gulf War, and the tragedy of the commons [13:30]
There are a number of examples of quick response manufacturing. One was called, in the mid-1990s, agile manufacturing. Some people at Lehigh University had convinced their congressman from Pennsylvania to put millions of dollars into the budget for agile manufacturing. This was in part because of rapid prototyping need that came out of the Defense Department in the First Gulf War. Some of you were probably born at the time of the First Gulf War but were still wearing diapers. Some of you weren't even born in the early '90s.
In the First Gulf War, we said, "We're going to kick Saddam Hussein out of Kuwait, we're going to mobilize, it's going to take us six months to get everything over there, and then we're going to attack." And we did. We got everything together in six months, attacked, and in like four or five days we wiped out Saddam Hussein's army, and stopped just short — George Bush senior stopped just short of wiping out Saddam Hussein because they knew if you wiped him out, the whole social infrastructure of Iraq would fall apart. And his son proved that, by wiping out Saddam Hussein, and Iraq's still sort of a social mess.
The Defense Department became very interested in rapid prototyping. How could you build something? At the time, we were all set for a war in Europe with tanks, or a war in Southeast Asia. We were not set for a war in the desert. The helicopters were failing because they would ingest dust and sand from the desert, and within three minutes they would just abrade away the whole engine. They had to develop things called engine particle separators and stuff. They worked around it, but they realized they were working very fast to solve problems. When you have a buildup that takes six months and a war that only lasts for five days, you don't have a lot of time to prototype or to change things. It took them six months to change things over for a desert war that they weren't prepared for. Fortunately, Saddam couldn't do anything about it in the six months.
That was rapid prototyping, and so these guys at Lehigh came up with the term agile manufacturing. They said, "It'd be wonderful if you could be more agile than your competition and make things more quickly." Then people took that and ended up calling it one-of-a-kind production. At the time, I said: I don't want to be able to make one-of-a-kind, because if I can make one-of-a-kind, so can my competitor in Thailand, and he probably has lower labor costs. I just want to be more agile than my competition. I want to be able to do things better than my competition. I don't want to be able to make things one of a kind.
One-of-a-kind is what should be called the tragedy of the commons. Does anyone know what the tragedy of the commons is? No one's ever heard of it? There was a 1968 paper in Science magazine by Garrett Hardin. He was big on overpopulation of the world — an ecologist, and also sort of a right-winger, which gave him some problems with his political ambitions. But the tragedy-of-the-commons phrase actually comes from an 1833 article written by a British author.
We have Boston Common here, right? Do you know why it's called Boston Common? The law back in those days was that farmers could come and graze their cows on the common. It's the common grazing area. The law still says you can graze your cow in Boston Common — they may now make you clean up what the cow leaves. Does anyone know why it's called Beacon Street in Boston? You know where Beacon Street runs from the common all the way through the Back Bay? Before they filled in the Back Bay, it was a bay, and Beacon Street was a causeway that you could walk your cow to the common on. When you came back at night, your cow had to wear a beacon — it was such a narrow causeway, someone else coming might be knocked into the bay if you didn't have a beacon to warn them. That's why it's called Beacon Street. You didn't know that? You've lived in Boston for months.
The tragedy of the commons is today, in sociology, when people act in their own self-interest to the detriment of the whole society. If some people go and overgraze the common, then there's nothing for the rest of the cows to eat. So it's the tragedy of the commons. We tend to overfish in the world — that's the tragedy of the commons. It's a buzzword for people acting in their own self-interest to the detriment of the rest of society. One-of-a-kind production, in my opinion, would be tragedy of the commons. Who can make money when the whole world's at one-of-a-kind production? It's like Uber and taxis. The Uber drivers don't make money. The managers at the top of Uber make all the money. I don't think the Uber drivers can even cover the cost of their automobile in the long term. That's a disaster waiting to happen. Although it's nice to have cheap rides in nice cars.
§5. 3D printing — ha [20:11]
3D printing. That's one of these things about quick response manufacturing. Why do I put "ha" in there? Anybody know why I think 3D printing is way oversold? If I wanted to make something like that turbine blade — assuming I could get all the properties — what do you think it would cost? If I wanted to make a titanium part about that size, what do you think it would cost in 3D printing today? About $10,000. And people say, "This is the new manufacturing miracle."
Now, it turns out that turbine blade is worth about five or six thousand. But that's because it's got properties that you can't get from 3D printing. So maybe not the best example. But people are talking — well, faculty in this department have raised $220 million for a startup firm that's using powder metallurgy for 3D printing, and they're going to make automotive parts. You've got kind of a little hurdle from $10,000 a pound to something that's automotive, which is going to be selling for on average about $2 a pound. You better have some really good ideas of how to reduce that. So what are they doing? They have multiple print heads. They have like twenty print heads. That could reduce the cost by a factor of ten. Ten into 10,000 and you're down to 1,000. You only need another factor of 500. So what's your next idea?
3D printing's not all it's made out to be in the popular press. And the tragedy is — big firms like General Electric have invested in part of this $220 million venture. The first aspects of 3D printing that were commercialized were medical devices, because people don't really care: you can sell a little part the size of a golf ball for $5,000 if it's going to help someone walk again. Medicine or health things are price insensitive.
Another thing people used to do early on — and I'm talking the early '90s — was Ely Sachs and Mike Cima here at MIT. Ely in mechanical and Mike Cima in materials, who basically took an inkjet printer and started injecting a glue and making a ceramic 3D part. That was the first time I ever heard the term 3D printing. It was 3D inkjet printing. They took a two-dimensional inkjet print head, modified it so they could put a slurry — actually an adhesive — and then they would put a bed of ceramic down, print a two-dimensional object, then move it, put another layer of loose powder down, print that, and they could build up things. Now there are all kinds of 3D printing. The second thing was pharmaceuticals.
You've got to get the right drug dose in that pill. Ninety-nine percent of that pill is just carrier. It's calcium carbonate, or milk of magnesia, or magnesium carbonate. It's just a carrier. The drug is only a few drops of something. And the 3D print head was great for putting the right amount of drug into that little pill rather than trying to mix them the way they used to. So there were some high-value-added parts for 3D printing. But now everybody's touting it as the manufacturing tool of the future because it's additive manufacturing rather than subtractive. You want to run a race? I'll take subtractive manufacturing any day over additive manufacturing. That's why I say "ha." You can find articles now where people are coming out and actually giving the cost on these things, and they are beyond belief. Actually, not beyond my belief — I've been working on similar things for years.
§6. TQM tools and the Ishikawa diagram story [24:54]
Total quality management tools. For coming to class, you get a free $5 book. [Tom distributes copies of the Memory Jogger.] It's called the Memory Jogger. They sell them for five bucks apiece. You can get a discount down to $4.25 if you buy 100. It's something you can put in your pocket. It tells you what a pie chart is, a Gantt chart, a histogram plot. These are some of the tools of total quality management. "A pocket guide of tools for continuous improvement and effective planning" is what it says on the cover.
It breaks things up into working with ideas, working with numbers, and working with teams. It has two pages on Cpk — we talked about Cpk yesterday, process capability and how it's calculated. It has a control chart in there. It doesn't call it a Shewhart chart, which historically was the original name, since Shewhart came up with it. Working with ideas — one example is a Gantt chart. Anyone who's worked in industry knows a Gantt chart, because you have certain tracks of things you have to do, and these are the times or dates you have to have them done by. You're smiling — you've seen them, right? Really sophisticated. Goes back to 1910. This is not exactly new. It's been around for a century.
There's the fishbone diagram, also known as the Ishikawa diagram — for the guy in Japan who said, "I can look at my management issues, raw materials, working conditions, technology, machine, and workers." So you've got management and labor, machines, input raw material. All these effects could be affecting the quality of your output. When you don't know what you're doing, you draw a fishbone diagram and try to isolate where the problem is going to be.
As with most things, I have a story. This story comes from what used to be called Detroit Diesel Allison. Now it's still in Indianapolis, but it's called Rolls-Royce aero engines. Rolls-Royce bought Detroit Diesel Allison from General Motors. They build aircraft engines and all kinds of other smaller engines for industrial use — typically turbine engines.
They had a particular problem. I can't tell you which engine it was, but they had thirteen unexplained failures, one or two of which actually killed a couple of people in an aircraft. The others, the aircraft was able to land before it crashed. They didn't really understand these failures. It was part of the turbine disc system. These thirteen failures had occurred over about a ten-year period — it wasn't something that was falling out of the air every week. They had made hundreds of thousands of these components and they'd only had these thirteen failures, but they were unexplained. That bothered the manufacturer of the aircraft who was buying these engines and putting them in their aircraft. It's just not a safe thing to have unexplained failures.
They put together what they called a blue ribbon team. They had a number of consultants from the outside, internal consultants from the aircraft manufacturer, internal consultants from Rolls-Royce. They brought us together for two days. They had a good-sized conference room, probably ten times the size of this room. They had charts all over the wall. One of them was a wall as big as this front wall with a huge Ishikawa diagram.
On the afternoon of the first day, they had someone scheduled to go over this Ishikawa diagram for an hour, to talk about all the fishbones on the diagram. It's sometimes called a cause-and-effect diagram. As this person was about to get up and be introduced, I raised my hand — I was one of about ten people on this panel. I said, "Can we not discuss the fishbone diagram?"
I said, "I know it's a useful tool when engineers are trying to convince management that they thought of every possibility. But if you're trying to figure out what the real problem is, it's just a waste of time. If you're really trying to get down to the root cause, why do you want to talk about a hundred possibilities? Why don't you use your expertise to focus in on the probabilities?" I fairly forcefully said this. I didn't say "you're wasting my time," but the manager at Rolls-Royce who was introducing this person was sort of taken back. It was about time for a break. He said, "Let's have a break. We're ready for our afternoon break and then we'll come back."
A bunch of engineers from Rolls-Royce came up and said, "We're so glad you said that." It's a good thing, when you don't know what you're doing, to just put up all the possibilities. It's called brainstorming sometimes. This is just one method of brainstorming. But if you're actually trying to focus in on a real answer, it's not really all that useful. We came back, and several of their people then felt empowered to go and tell their management bosses that they agreed it wasn't really useful to spend an hour and a half on it. I would have fallen asleep — it was just after lunch. It's just deadly. Some of these things have uses, but they're not universally useful. You have to think about when they apply and when they don't.
§7. Is TQM for real? Schrödinger's cat [32:16]
So we come up again with: is TQM for real? At one level, TQM is a series of buzzwords. If you are getting the impression it's a series of buzzwords, it's probably a waste of time. At another level, it's a set of problem-solving methods and tools that can be an important part of engineering. Brainstorming is an important part of engineering when people share ideas. TQM really involves people and management of people. Ninety-five percent of all problems in industry are people problems. They're not technical problems. At the root of the problem, it's people.
Student: [indicates agreement]
You work in industry? You agree with that statement? You do. Have you worked in industry? No. You agree with that statement that it's people problems? Have you worked in industry? Our videographer — have you worked in industry? You agree that people are the problem? Okay.
The reason CEOs like Jack Welch and Ray Stata and these other people were interested in TQM — it basically was trying to teach the principles of Douglas McGregor, theory X and theory Y. You really want to empower your workforce to do continuous improvement. I can string together all the buzzwords of TQM, but it really gets down to getting everyone united in working out the problem themselves, rather than having someone try to pass it down from on high and order them to do things.
What I've tried to present here on TQM is sort of a conundrum. Is it good or is it bad? Is it for real? A quote I learned in the early days of the LFM program was from F. Scott Fitzgerald: "The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time and still retain the ability to function." Is TQM good or is TQM bad? The answer is actually both. It's a little bit of both. How are you going to deal with it when you go out into the real world and you start getting some business school buzzword guru or some consultant start buzzwording at you — can you bring it back to something that's real and tangible and has real engineering content?
The person who quoted this first part to me never quoted the second part, and it was a number of years before I read the second part, because I didn't fully comprehend this until I read what F. Scott Fitzgerald said in this interview following the first part: "One should, for example, be able to see that things are hopeless and yet be determined to make them over otherwise." If it's hopeless, why would you work on it? You'll hit hopeless problems all the time. I talked about the LFM student who went into the Mad Hatter company. The morale was terrible. It was hopeless. By taking a different tack — buying donuts for the hourly workers — she was able to get things done that the whole management team at this huge corporation had not been able to do for years. Why? Because they had the theory X attitude that the workers are bad, they don't really want to work. She started treating them with some respect and found that they would respond.
What I often say is, I don't know anyone who goes home at the end of the day saying, "I did a really lousy job today and I'm proud of it." They go home saying, "I did a really lousy job today at work because they wouldn't let me do anything else. I was trying to tell them how we could do it better, but they wouldn't let me." They shift the blame to someone else rather than taking the responsibility of forcing it on the supervisor that's forcing it on them.
So, is it good or is it bad? Here's a cartoon I had on my desk somewhere about your cat, Mr. Schrödinger: "I have good news and bad news." TQM is both good and bad at the same time. That's what Schrödinger's cat is all about. He's alive and he's dead at the same time. Just keep the opposing ideas in your mind at the same time and you'll be able to deal with it.
§8. Sioux City and jet engine reliability [37:25]
This is only my fifth lecture, but I'm gone all next week. So Brian Homan [Hohmann] will be lecturing on the example of the Sioux City, Iowa crash and the problems of inspection and quality control. Anybody know about Sioux City, Iowa? It was probably back in the 1980s, so only a couple of us were alive then. A disc in a GE engine blew up. It's McDonnell Douglas, because they had both engines in the back on the tail. It wiped out both engines when it went.
All of a sudden, the pilot had to do a glide into Sioux City, Iowa, airport. He did it successfully — sort of like the guy who landed the US Airways airplane on the Hudson River. They made a movie out of that. Sioux City, Iowa was before. It turns out it was a titanium nitride inclusion from the casting process that caused a fatigue crack in the titanium disc. Brian has a lot of stories about this, including how it caused General Electric to redo a lot of their quality control metrics. You have a serious problem when you get to five and six sigma. How do you fix the problem and get to a higher level of reliability?
Does anybody know the typical reliability of a commercial jet engine? They're like 99.44 percent uptime. That .56 percent is actually fairly significant, because it means you'll shut down the engine one time for every 200 hours of operating. How many people have ever been on airplanes for more than 200 hours? The odds are they shut down the engine for cause on one of those aircraft you were on. They don't usually come over the announcer and say, "Well, we just shut down one of our four engines because it doesn't work" — some people wouldn't be real happy. But it happens all the time. The numbers I saw twenty years ago — and they were GE numbers — were 99.37. This is not something they advertise a lot, because some people would be like me, and they take the inverse of 99.37 and say, "Oh, actually, that's one time out of 300 hours." I've been on more than 300 hours' worth of airplanes in my life.
§9. Leadership, management, and education [40:30]
I wanted to talk a little bit about what this course is all about. Brian will talk about Sioux City, Iowa one day next week, and he's got to get together with Steve Lyons, because Steve Lyons will be finishing up his lectures. Dr. Belmar finishes his module of six tomorrow. Mr. Lyons will do a full double module of twelve. If you're attending all the live ones, we're going all the way through March 13th or 14th.
I gave you at the beginning this article on leadership, management, and education at MIT. When I was in LFM, I was the co-director, and it was Leaders for Manufacturing. Now it's Leaders for Global Operations. And what do these leaders mean? I thought about it from time to time, put together some thoughts, wrote this article called "Leadership, Management and Education," and put it in the faculty newsletter.
You can go back and read this. I wanted to get a pithy beginning: "The world looks to MIT for leadership, and this leadership is not limited to science and technology." A kid drank himself to death and they gave his parents millions of dollars. He'd only been at MIT for six weeks. He drank a fifth of Southern Comfort in two hours, choked on his own vomit, and all of a sudden this place was in an uproar. We were on the front page of the New York Times, the Washington Post. How could someone from MIT, the best and the brightest, drink themselves to death? Well, your parents taught you how to drink before you got here for six weeks. We didn't teach you. But nonetheless, MIT was blamed for it. They paid millions of dollars to his parents. I'm sorry he died, but gifted people aren't supposed to do this.
The expectations of MIT being better than average are not new. In 1911, Thomas Alva Edison was quoted: "There is no question but the Massachusetts Institute of Technology is the best technical school in the country. I have found the graduates of Tech to have a better, more practical, more usable knowledge as a class than the graduates of any other school in the country. The salvation of America lies in the Massachusetts Institute of Technology." Doesn't that sound good? It's worth quoting. You don't engrave this on the Building 7 dome or something.
Sitting in a bunch of boring meetings at LFM, I had read one night — or one person had given a talk the night before — that Peter Drucker, the management guru, had talked about how a leader gets the right things done. I started thinking, what else can I write down about what leaders do? They get the right things done. Back in the 1980s, I'd determined that half of what I did didn't need to be done. I just didn't know which half. You get all these tasks to do, and if you just ignore half of them, they would go away — but you don't know which half is going to go away. So you have to pay attention to most of them. A leader can sometimes figure out what to do and what not to do.
A leader does more than is required. I've worked with a number of people, and they want to know what the bare minimum is that they can get by with. That's fine for them, but they're not going to progress if they're doing the bare minimum. They balance professional and personal responsibilities.
When I was 43, I was put on the board of a manufacturing company that had $500 million in sales. They'd been around since 1843. Company called Nashua Corporation up in Nashua, New Hampshire. They started out making playing cards — so all the cowboys out in the west could have something to do at night. They'd play cards, and Nashua made them. Nashua got into a number of things. They hired a new CEO. The old CEO had been there for twenty-five years. He had helped save the company from bankruptcy. He was a very good guy — of course he was a good guy, he's the one who put me on the board. We had a problem after one year with the new CEO, and it was partly California culture coming into New England culture, and they're not the same.
Charlie said, "I interviewed him, but I should have interviewed his spouse also." Because she still wanted to be in California. What I've seen a number of times is if someone's not doing well at home, they're not going to do well at work. They're worried about their children, their spouse, their finances. That keeps them from focusing on the job at hand. So people have to be able to balance both professional and personal responsibilities. If they can't, they're just not going to be able to help other people.
They respect the contributions of everyone. I get in fairly early in the morning, and I used to park right here by Building 10 before they started building the new Nano Center. The nighttime custodians would check out at 7:00 in the morning, so I'd be walking past them at different times, and I got to know some of them. When I became department head, I took the Chipman Room — a different Chipman Room than this one, with a little kitchenette with a stove. Our administrative officer was a guy named Joe Dorsey. I said, "Joe, this year at Christmas, let's do a breakfast for the nighttime custodians." These were hard-working people. I knew I couldn't do it from 7:00 to 8:00 because most of them left this job to go to their next job. They were working two jobs to make ends meet.
I got permission from Donna — the chief of physical facilities — and said, "Look, we want to do a breakfast for the nighttime custodians. Can you let them off at 6:00 and they can come and we'll feed them breakfast?" We fixed them scrambled eggs and French toast. I got up as department head, with the administrative officer, and said, "We appreciate what you've done this past year." They were so overjoyed. We had the best custodial service in the institute for the next few years. Every year we would have this Christmas breakfast.
You know what the rest of the faculty thought? "Why are you doing this?" They couldn't figure out why. Most of them that know about it still can't figure out why I did this. But we had the best custodial service. Because all I did was show respect for people who deserve a little respect. The first things you would notice if these people went on strike would be the restrooms. The second thing you'd notice is — if the support staff is gone — there's no food for lunch. So you have to respect the contributions of everyone. This place would not work. No operation will work without everyone's contribution.
They also contribute to the community as a whole. They're going to provide some service because they've got their own life balanced. They're doing more than required, and more than required involves also the broader community. The most important thing is perhaps they're willing to follow others when they're not leading. They don't have to be the boss. They can change things from the lower levels by just being a good example.
So what do you think this course was about? I asked two of you ahead of time. It was about leadership. TQM is a form of teaching leadership and helping people work together, respecting people and knowing how to work together to make the whole system work — realizing that the ideas of the whole are better than the ideas of a single person. I hope that's what you can actually take away from the five hours of lecture here. Not just the stories, but the fact that everybody's important. Thank you. If you have other questions about the course, I'll be happy to entertain them. Otherwise, see you around.