How GM's Culture Will Prove Its Demise
General Motors will soon be bankrupt. I'm not saying this out of spite, out of schadenfreude, out of a need to lash out at the auto behemoth. Rather, my judgment stems from a realization that the monster has become not only too big to turn its business practices around, but is further infected with a corporate culture that lacks the initiative to even attempt such a reversal of practices and fortunes. Anyone can see this to be the case. All you need to understand are the concepts of corporate culture and how they differ from business to business, from culture to culture. Right now, the biggest three auto making countries are the US, Japan and Germany. (Other car making countries like Korea, China and India are growing in importance, but their products and histories are not really available for me to judge, so I'll stick with the big three as I tick off the elements of each that give my argument some weight.)
Let's start briefly with Germany. Years ago, I worked with a deckhand/diesel mechanic named Jack. How good a diesel mechanic was he? He drove an old Mercedes diesel he had kept running for all but the first 100,000 of the 600,000 miles on the odometer, and the beast still ran like a top. It was a work car, to be sure. Jack kept the exterior paint minimally polished, and had replaced the back seats with a mechanic's tool box the size of a small fridge bolted to the floor panels.
One story Jack told fascinated me. As a young man, he had been accepted to apply at a diesel mechanic training school somewhere in California, a school run by a German diesel firm. The German-operated part of the story proved the most important element. To even get accepted to apply for this school, applicants had to pass a basic test on parts and maintenance knowledge and have some mechanical work experience. Once in, they were gathered in a workshop for an "entry test" of those skills. Jack described this workshop as an old-world machine shop, filled with chunks of metals and the tools to shape them -- all hand tools. Files, vices, crank drills, saws, thread tap and die tools, all lay about in abundance, all harkening back to the era before the electric motor made many of these tools easier and faster to use. To guide these hand tools, applicants had access to a vast array of calipers, straight edges, scoring points, compasses, levels, plumb bobs, angle tools, every manner of analytic device ready and waiting for knowledgeable hands and eyes.
Then came the test. Applicants were told to bring a lunch, that the test could take all day. At eight in the morning sharp, the examiner started his spiel about how the school trained only the best, and that the work they did in the work shop would be graded by its quality. Each applicant was given a hunk of scrap metal. The test: using as much time as needed, drill a hole in their scrap at a specified angle and gauge, and tap it with a specified thread count. Each applicant had different a angle, depth, gauge and thread pitch to achieve. Go.
Jack and his co-applicants naturally went about the slow and laborious task of fashioning individual drill guides to help them properly drill and tap their hole. Get a scrap piece of metal, drill a proper hole perpendicular to the surfaces, saw it to the proper angle, devise a vise to set it to your test piece (or, as Jack told me, get another scrap piece for a dry run first), drill and tap your hole, check it, and turn in your work.
Anyone who has ever been in a testing room can predict what happened next. The applicants performed their test task with varying levels of speed and acumen. Some whizzed through the test in just a few hours, turned in their work, and left. Jack was a deliberate and patient guy, and took a few more hours than most. He worried that he wouldn't finish in time, but turned in his work just before 4:00 pm and left. By that time, only he and four others remained in the testing room.
Later that night, he got a call congratulating him on a test well done. He was told to start classes the following day. On that first day of class, though, he got a surprise. About half of the applicants from the previous day were there, but nobody who turned in their work before noon was present. These instructors followed a work ethic very different, almost foreign to the US students; methodical, careful work was to be rewarded. Nobody was rewarded for their natural talent when it came to getting the job done quickly. Why? To these master machinists and mechanics, one took all the time one needed to do a job correctly.
Jack told me something else he learned. To these German professionals, a properly trained and experienced mechanic could solve just about every problem . . . except problems to which they weren't familiar. In that case, no amount of introspection and examination would help simply because they literally wouldn't see the problem, and would try everything with which they were experienced before considering something new.
I gave you that lengthy introduction to the German engineering philosophy to demonstrate the power culture has over all of us. Another friend told me about the BMW TurboSteamer Engine, something he read about in Car & Driver. (Sadly, the original Car & Driver article is not available online, so I am forced to relate the hearsay story from my friend.) Essentially, the engineers at the Bavarian Motor Werks held the concept of the gasoline-electric hybrid in disdain. Why? Using an electric motor to aid propulsion and efficiency was, well, cheating. A proper gasoline design engineer should work to remove the inefficiencies from the existing propulsion system before resorting to gadgetry outside the automotive traditions already so well established. Since one of the most obvious inefficiencies of any internal combustion engine is a vast amount of heat shot out the tailpipe, BMW developed the TurboSteamer in 2005 to recapture that heat and convert it into additional torque. According to the the Wiki entry, the system increased the fuel efficiency at higher speeds by about 15%. Not bad.
(Of course, my friend and I eschewed the engineer's philosophical disdain and started dreaming What If. We immediately realized that the system would be much better suited not for luxury cars, but for heavy equipment propulsion like railroad locomotives. Big rail engines can run almost constantly for days instead of hours, and pull too much mass to make a bulky battery efficient in any way whatsoever. Extracting that 15% extra power per gallon from a train would add up to fuel cost savings in the hundreds of thousands of gallons per year. Furthermore, most rail engines are already diesel-electric hybrids, using the diesel generators to power the electric drive motors. A TurboSteamer concept here would make more electricity more efficiently than more torque, since the torque would likely need to be geared to the specific driving conditions. But I digress. . . .)
There are two salient facts to note here. The German engineers (at first) held the gasoline-electric hybrid idea in contempt, developing the TurboSteamer as an alternative. That we've discussed. Second, that propulsion system has not yet left the test rack at the lab. Let's compare that to another design philosophy, one developed by Toyota, developer and builder of the Prius.
Toyota runs by a philosophy known as The Toyota Way, 14 Principles that guide the company:
The 14 principles of The Toyota Way are organized in four sections: 1) Long-Term Philosophy, 2) The Right Process Will Produce the Right Results, 3) Add Value to the Organization by Developing Your People, and 4) Continuously Solving Root Problems Drives Organizational Learning.
Core to this philosophy and its 14 organizing principles lies the concept of kaizen, "the process of continuous improvement." Instead of (like BMW) spending years perfecting a system before it hits the road, Toyota might build and test a system to make sure it should function, review the problems customers experience, and continuously tweak the design to fix those problems. After all, there is only so much the lab test bench can tell even the most gifted engineer. For example, consider the problem the first batch of Priuses experienced in slippery conditions: They couldn't move. The varying torque profiles of the gasoline and electric motors worked pretty well on an average pavement, but once the snow fell or the ice formed, the software systems that balanced the two forces got all out of whack, tending to give the wheel a slight lurch then stop and try again, creating a pulsing effect that resulted in a rhythmic shuddering -- but no forward movement. (Though I've never driven a Prius Gen I, this is something I've experienced it in a similar drive system.) It took owners stranded in snowy areas to discover this problem before the engineers could fix it in later models.
A few other Way concepts caught my eye. Take Principle 3: Use "pull" systems to avoid overproduction. Essentially, demand downstream on the production line dictates production upstream. This means all sections of the assembly must know what is needed upstream to work efficiently. This caught my eye because of my earlier reading. In The Evolution of Useful Things, Henry Petroski noted that "Oriental hand saws . . . cut on the pull stroke (as opposed to Western saws, which cut on the push stroke) . . ." (Petroski, Vintage Books, 1994, p. 119). This corresponded to an article on hand wood planers (the source long forgotten, sorry) which also noted that Japanese planers are always pulled toward the operator due to a strong cultural bias against "pushing" work. It seems this bias to pulling is still alive and well in Japan, steeped not just into the craftsmen's approach to tools but into the corporate cultures that employ them.
I've worked on assembly lines before, and I can tell the differences between the US and Asian philosophies. In the US, you do your work. You don't slack off. If you finish your work early, you (as a worker) might get to go home. If you fall behind, you'd better pick up the pace. From what I've read about the Toyota system, "your work" is dependent on what happens to the work later in the process. This means that you might get to slow down your production if less production is needed. For example, take Principle 4: Level out the workload . . . . (Work like the tortoise, not the hare). Why rush one section of the assembly if the rush can't be used downstream? What a great idea. The worker can use this extra time in production to think about ways to improve his workplace efficiencies. Even better, should the worker have an idea, management is tasked to actually listen according to Principle 12 Part 4: "Take full advantage of the wisdom and experiences of others to send, gather or discuss information."
Most managers I've had wouldn't listen to my utterances unless I prefaced my comments with "Fire!"
For another example of how the "pulling" philosophy works at Toyota, look at the Principle 8 description: "Technology is pulled by manufacturing, not pushed to manufacturing" (Emphasis in the original). This might seem counter-intuitive given the additional complexity a hybrid gasoline/electric drive presents to the finished car; but bear with me. I think this is relevant. First of all, the basic components of Toyota's hybrid drive are well-known, from the gasoline engine, to the electric motor, to the differential gear that combines the two forces. After all, just about every car on the road uses a similar limited slip differential to balance the torque to the drive wheels. What is innovative in the Prius is the computer hardware and software and the sensors that dictate and balance the various inputs. That's where the Prius' true complexity lies, in the brains driving the system rather than the system itself.
According to this Wired article detailing the General Motors chop shop, this rationale wasn't good enough for General Motors. The story describes the process every car maker undertakes to examine the competition by taking cars from competitors apart piece by piece and figuring out how they were assembled. This process can reveal new technology that can help future models.
A full teardown takes about six weeks. First, mechanics measure the vehicle with a device called a 3-D vector arm, taking all of the car's inner and outer dimensions, like the bumper height and the distance from the driver's eyes to the steering wheel. They create a digital blueprint, then they disassemble the car. Each part gets named, weighed, and labeled with a number. Cost estimators gauge the price of every one, not only to determine what competitors spend but also to pressure GM's own suppliers. "We know a certain kind of plastic costs x per kilo," explains staff project engineer Craig Duncan, a round man in standard GM dress of khakis and a polo shirt. "So we know the mass of the part, what the labor rate is, and what the shipping costs are, and we start adding up all the puzzle pieces. It's a scientific way of being much more aggressive with our suppliers to push the cost down." Finally, all the information is entered into a database for GM engineers puzzling out new car designs.
What was especially interesting to me in this article was the frank discussion of hybrid technology in the preceding paragraph:
Every company wants to know exactly how its competitors' cars are put together and how much they cost to make so it can learn how to save money on parts, shed weight, and improve its manufacturing methods. Even more important, teardowns help executives make long-term strategic decisions. A teardown of the 2004 Prius two years ago helped sour GM on hybrid technology. The company is slowly rolling out hybrid trucks and buses, but it's focusing its innovation efforts on fuel cells. (Emphasis mine.)
What catastrophic information did the Prius teardown reveal to the sages at GM?
A dissected hybrid lies on display in the back of the VABA, next to the remains of a Chevy Malibu. The Prius demystified comes down to this: It has 1,432 propulsion parts (the Malibu has 822). It has two electric motors, plus lots of software and finely machined gears to transport power to the wheels. That makes the Prius very expensive to design and build. (Once again, emphasis mine.)
Oh, waaaaaaa.
To Toyota, this decision to try something new in the hybrid design was nothing new. Remember, they fully expect to improve the design slowly over time, gradually improving the Prius or its successors. These first few model years are going to prove the test bed for the more fully realized hybrids of the future. That's kaizen in action. If they have to take a loss on those first few model years, so be it. At Toyota, that's how progress is made. Based upon their current market standing, I'd have to say their philosophy seems to be working.
At GM, it seems, progress has to be made in the book keeper's ledgers before it can ever leave the factory floor.
That part counting attitude shows up in the only GM hybrid design on the road (to my knowledge), the Allison Electric Drive system (Allison is a GM subsidiary that focuses primarily on truck transmissions). According to this article:
Allison Transmission's electric-vehicle think tank in northeast Indianapolis could start clearing urban smog by mid-decade with its electric solution. Replacing the hefty diesel engine with a slimmer model hooked to sizable electric motors would power the 20-ton buses just like environmentally friendly cars such as Toyota's 52-mile-per-gallon Prius. Buses equipped with Allison's diesel-electric system won't match Prius' mileage, but bus fuel economy could improve 60 percent, reducing smog and related respiratory ailments as buses burn fewer gallons of diesel oil. (I'm still doing that emphasizing.)
I drive those diesel hybrids at work. The fuel savings estimated in that article proved not so great. No one would give me specifics on the record, but the Allison system did not improve fuel economy that much. It's safe to say the number above is off the actual fuel savings percentage by an order of magnitude, at least with the buses of our fleet. I'd be very curious to learn if the newer generations of Allison drives have improved much at all.
Fuel savings be damned. Let's talk about one aspect of GM's focus that drove their research dollars in absolutely the wrong direction. Again from the Wired article, remember that little blurb about "focusing its innovation efforts on fuel cells?"
In the back of the teardown building, alongside the dissected Prius and Malibu, lie the parts of a GM demonstration vehicle powered by hydrogen fuel cells. Although the vehicle has more total parts than the Prius, almost none of them move, eliminating many of the finely machined gears and engine components of the traditional auto. The tables bearing the drive trains of the Prius and Malibu are laden with parts; those for the fuel cell car are nearly empty. The three teardowns tell the story of GM's plans: Go directly from gasoline to fuel cells with a mere nod to hybrid tech. . . .
It seems like a brilliant move. . . . But GM's critics aren't so sure, calling the company's fuel cell talk "defensive PR" because it's so far behind Toyota on hybrids. Auto industry consultant Maryann Keller points out that fuel cells are every bit as complicated as hybrids, if not more. "You're talking about a completely different propulsion system!" she says. "No one can even repair or drive one!" She argues that, instead, GM needs to follow intermediate steps. "What technology doesn't become cheaper and easier to build over time? . . . Hybrids are complex and expensive, but you'll never figure out how to make them better and cheaper until you put the technology in real-world cars." (I emphasized.)
Many may not realize the sheer ironic gravity of that last emphasized statement. Please allow me to illustrate.
Yes, GM is supposedly pursuing fuel cell technology (or was, back when they still had some money). What is a "hydrogen fuel cell powered car?" The fuel cell takes hydrogen from an onboard storage tank, combines it (usually) with atmospheric oxygen to produce electricity, and exhausts the resulting water vapor. The electricity produced by this cell drives one or more electric motors to make the car go zoom. So, to follow the sound logical argument Ms. Keller notes in her last quoted statement, GM probably should build electric cars to improve that technology. With me so far?
Alright. GM did build electrics. They produced probably the best US-built on-road electric car, the EV-1. People loved them. People had to pass a test just to lease them. Not "buy," mind you; lease. That test probed into all kinds of personal information only minutely relevant to most car buyers, asking applicants to state their incomes, verify they had a private house with a garage, and own up to all kinds of past purchases like (according to one question from the questionaire) a Betamax when they had been available. Supposedly this showed you to be an early adopter willing to take risks with a new technology. After all that, the "winning" drivers had to cough up about $3,000 to install a charger in their garage -- installing on-street chargers other cars could use were specifically not an option. Only then were they given the option of paying $450 a month for three years to drive the cars.
And once that lease expired, GM took the cars back.
No purchase option was ever made to any of the drivers, even when drivers literally offered a million dollars for as many as that would buy. (In fact, some cars were taken back early, the result of a recall. For their $450/month, those lucky few were given Geo Metros to drive through the rest of their lease. The problem prompting the recall was never fixed.)
After all that, what did GM do with its successful fleet of electrics, cars that could have yielded a goldmine of development data as it rolled through the roads of California and Arizona? 40 were donated to universities and museums, but only after the drive components were removed and the institutions signed a legally-binding promise never to get the cars running, ever again. Only one EV-1 remains intact, the one on display at the Smithsonian. The Smithsonian has a policy on authenticity and would not accept a "disabled" car . . . though the one it displays does have a dead battery pack.
And the rest? What was to be their fate? Take a peek.
They were hauled to the desert and crushed.
"Hybrids are complex and expensive, but you'll never figure out how to make them better and cheaper until you put the technology in real-world cars." Oh, and we should add, ". . . and you're willing to learn the lessons these cars can teach you."
Based on the Wired article, the movie (a story which I followed from the owners' perspective online as it was unfolding), and just about every other horror story I've heard about the GM "design" process, the company is doomed. The only kaizen they have available to them applies only to the bookkeeping department. Company designers and engineers have this Sword of Damocles called The Bottom Line dangling over every move they make, over every innovation they suggest or pursue. Yes, Rick Wagoner was asked to leave the company; but he was replaced with another executive brought up through the ranks in the same poisonous miasma of Profit Über Alles that spawned Wagoner and all his predecessors, the folks that cumulatively got GM into this escalating mess that will soon prove their demise. Oh, but don't just take my word for it. Evidence for this self-defeating culture abounds. For a look at this short-term profit culture at work from a very different angle, see bradhicks's excellent piece on the short-term gains the company got on worker pay raises verses retiree benefits that in the long run ultimately crushed all of their profits to a flat pulpy mess.
That's the insidious problem with culture, be it one shared by groups in a company or a country; one does not succeed unless one at least outwardly accepts the "way things are done around here." After a while one accepts this Way and even subconsciously absorbs it as Normal, wrong though it may be. Maybe it's in the way you develop your company's products. Maybe it's as simple as the words everyone uses so often around the office that don't exist elsewhere. Whatever it is, we human beings will overtly or subtly (depending upon the humans involved) enforce the worldview that the group uses to pursue its ends.
Worldviews don't die easily . . . at least not as easily as the companies and peoples that embrace them. Financial and physical realities have a way of pointing out the strengths, blind spots and inflexibilities in worldviews large and small. Winning cultural strategies can lead to success; losing ones lead only to death.
Addendum, Earth Day, 2009: I added the detail about the EV-1, something I had forgotten to add in the original draft.
Let's start briefly with Germany. Years ago, I worked with a deckhand/diesel mechanic named Jack. How good a diesel mechanic was he? He drove an old Mercedes diesel he had kept running for all but the first 100,000 of the 600,000 miles on the odometer, and the beast still ran like a top. It was a work car, to be sure. Jack kept the exterior paint minimally polished, and had replaced the back seats with a mechanic's tool box the size of a small fridge bolted to the floor panels.
One story Jack told fascinated me. As a young man, he had been accepted to apply at a diesel mechanic training school somewhere in California, a school run by a German diesel firm. The German-operated part of the story proved the most important element. To even get accepted to apply for this school, applicants had to pass a basic test on parts and maintenance knowledge and have some mechanical work experience. Once in, they were gathered in a workshop for an "entry test" of those skills. Jack described this workshop as an old-world machine shop, filled with chunks of metals and the tools to shape them -- all hand tools. Files, vices, crank drills, saws, thread tap and die tools, all lay about in abundance, all harkening back to the era before the electric motor made many of these tools easier and faster to use. To guide these hand tools, applicants had access to a vast array of calipers, straight edges, scoring points, compasses, levels, plumb bobs, angle tools, every manner of analytic device ready and waiting for knowledgeable hands and eyes.
Then came the test. Applicants were told to bring a lunch, that the test could take all day. At eight in the morning sharp, the examiner started his spiel about how the school trained only the best, and that the work they did in the work shop would be graded by its quality. Each applicant was given a hunk of scrap metal. The test: using as much time as needed, drill a hole in their scrap at a specified angle and gauge, and tap it with a specified thread count. Each applicant had different a angle, depth, gauge and thread pitch to achieve. Go.
Jack and his co-applicants naturally went about the slow and laborious task of fashioning individual drill guides to help them properly drill and tap their hole. Get a scrap piece of metal, drill a proper hole perpendicular to the surfaces, saw it to the proper angle, devise a vise to set it to your test piece (or, as Jack told me, get another scrap piece for a dry run first), drill and tap your hole, check it, and turn in your work.
Anyone who has ever been in a testing room can predict what happened next. The applicants performed their test task with varying levels of speed and acumen. Some whizzed through the test in just a few hours, turned in their work, and left. Jack was a deliberate and patient guy, and took a few more hours than most. He worried that he wouldn't finish in time, but turned in his work just before 4:00 pm and left. By that time, only he and four others remained in the testing room.
Later that night, he got a call congratulating him on a test well done. He was told to start classes the following day. On that first day of class, though, he got a surprise. About half of the applicants from the previous day were there, but nobody who turned in their work before noon was present. These instructors followed a work ethic very different, almost foreign to the US students; methodical, careful work was to be rewarded. Nobody was rewarded for their natural talent when it came to getting the job done quickly. Why? To these master machinists and mechanics, one took all the time one needed to do a job correctly.
Jack told me something else he learned. To these German professionals, a properly trained and experienced mechanic could solve just about every problem . . . except problems to which they weren't familiar. In that case, no amount of introspection and examination would help simply because they literally wouldn't see the problem, and would try everything with which they were experienced before considering something new.
I gave you that lengthy introduction to the German engineering philosophy to demonstrate the power culture has over all of us. Another friend told me about the BMW TurboSteamer Engine, something he read about in Car & Driver. (Sadly, the original Car & Driver article is not available online, so I am forced to relate the hearsay story from my friend.) Essentially, the engineers at the Bavarian Motor Werks held the concept of the gasoline-electric hybrid in disdain. Why? Using an electric motor to aid propulsion and efficiency was, well, cheating. A proper gasoline design engineer should work to remove the inefficiencies from the existing propulsion system before resorting to gadgetry outside the automotive traditions already so well established. Since one of the most obvious inefficiencies of any internal combustion engine is a vast amount of heat shot out the tailpipe, BMW developed the TurboSteamer in 2005 to recapture that heat and convert it into additional torque. According to the the Wiki entry, the system increased the fuel efficiency at higher speeds by about 15%. Not bad.
(Of course, my friend and I eschewed the engineer's philosophical disdain and started dreaming What If. We immediately realized that the system would be much better suited not for luxury cars, but for heavy equipment propulsion like railroad locomotives. Big rail engines can run almost constantly for days instead of hours, and pull too much mass to make a bulky battery efficient in any way whatsoever. Extracting that 15% extra power per gallon from a train would add up to fuel cost savings in the hundreds of thousands of gallons per year. Furthermore, most rail engines are already diesel-electric hybrids, using the diesel generators to power the electric drive motors. A TurboSteamer concept here would make more electricity more efficiently than more torque, since the torque would likely need to be geared to the specific driving conditions. But I digress. . . .)
There are two salient facts to note here. The German engineers (at first) held the gasoline-electric hybrid idea in contempt, developing the TurboSteamer as an alternative. That we've discussed. Second, that propulsion system has not yet left the test rack at the lab. Let's compare that to another design philosophy, one developed by Toyota, developer and builder of the Prius.
Toyota runs by a philosophy known as The Toyota Way, 14 Principles that guide the company:
The 14 principles of The Toyota Way are organized in four sections: 1) Long-Term Philosophy, 2) The Right Process Will Produce the Right Results, 3) Add Value to the Organization by Developing Your People, and 4) Continuously Solving Root Problems Drives Organizational Learning.
Core to this philosophy and its 14 organizing principles lies the concept of kaizen, "the process of continuous improvement." Instead of (like BMW) spending years perfecting a system before it hits the road, Toyota might build and test a system to make sure it should function, review the problems customers experience, and continuously tweak the design to fix those problems. After all, there is only so much the lab test bench can tell even the most gifted engineer. For example, consider the problem the first batch of Priuses experienced in slippery conditions: They couldn't move. The varying torque profiles of the gasoline and electric motors worked pretty well on an average pavement, but once the snow fell or the ice formed, the software systems that balanced the two forces got all out of whack, tending to give the wheel a slight lurch then stop and try again, creating a pulsing effect that resulted in a rhythmic shuddering -- but no forward movement. (Though I've never driven a Prius Gen I, this is something I've experienced it in a similar drive system.) It took owners stranded in snowy areas to discover this problem before the engineers could fix it in later models.
A few other Way concepts caught my eye. Take Principle 3: Use "pull" systems to avoid overproduction. Essentially, demand downstream on the production line dictates production upstream. This means all sections of the assembly must know what is needed upstream to work efficiently. This caught my eye because of my earlier reading. In The Evolution of Useful Things, Henry Petroski noted that "Oriental hand saws . . . cut on the pull stroke (as opposed to Western saws, which cut on the push stroke) . . ." (Petroski, Vintage Books, 1994, p. 119). This corresponded to an article on hand wood planers (the source long forgotten, sorry) which also noted that Japanese planers are always pulled toward the operator due to a strong cultural bias against "pushing" work. It seems this bias to pulling is still alive and well in Japan, steeped not just into the craftsmen's approach to tools but into the corporate cultures that employ them.
I've worked on assembly lines before, and I can tell the differences between the US and Asian philosophies. In the US, you do your work. You don't slack off. If you finish your work early, you (as a worker) might get to go home. If you fall behind, you'd better pick up the pace. From what I've read about the Toyota system, "your work" is dependent on what happens to the work later in the process. This means that you might get to slow down your production if less production is needed. For example, take Principle 4: Level out the workload . . . . (Work like the tortoise, not the hare). Why rush one section of the assembly if the rush can't be used downstream? What a great idea. The worker can use this extra time in production to think about ways to improve his workplace efficiencies. Even better, should the worker have an idea, management is tasked to actually listen according to Principle 12 Part 4: "Take full advantage of the wisdom and experiences of others to send, gather or discuss information."
Most managers I've had wouldn't listen to my utterances unless I prefaced my comments with "Fire!"
For another example of how the "pulling" philosophy works at Toyota, look at the Principle 8 description: "Technology is pulled by manufacturing, not pushed to manufacturing" (Emphasis in the original). This might seem counter-intuitive given the additional complexity a hybrid gasoline/electric drive presents to the finished car; but bear with me. I think this is relevant. First of all, the basic components of Toyota's hybrid drive are well-known, from the gasoline engine, to the electric motor, to the differential gear that combines the two forces. After all, just about every car on the road uses a similar limited slip differential to balance the torque to the drive wheels. What is innovative in the Prius is the computer hardware and software and the sensors that dictate and balance the various inputs. That's where the Prius' true complexity lies, in the brains driving the system rather than the system itself.
According to this Wired article detailing the General Motors chop shop, this rationale wasn't good enough for General Motors. The story describes the process every car maker undertakes to examine the competition by taking cars from competitors apart piece by piece and figuring out how they were assembled. This process can reveal new technology that can help future models.
A full teardown takes about six weeks. First, mechanics measure the vehicle with a device called a 3-D vector arm, taking all of the car's inner and outer dimensions, like the bumper height and the distance from the driver's eyes to the steering wheel. They create a digital blueprint, then they disassemble the car. Each part gets named, weighed, and labeled with a number. Cost estimators gauge the price of every one, not only to determine what competitors spend but also to pressure GM's own suppliers. "We know a certain kind of plastic costs x per kilo," explains staff project engineer Craig Duncan, a round man in standard GM dress of khakis and a polo shirt. "So we know the mass of the part, what the labor rate is, and what the shipping costs are, and we start adding up all the puzzle pieces. It's a scientific way of being much more aggressive with our suppliers to push the cost down." Finally, all the information is entered into a database for GM engineers puzzling out new car designs.
What was especially interesting to me in this article was the frank discussion of hybrid technology in the preceding paragraph:
Every company wants to know exactly how its competitors' cars are put together and how much they cost to make so it can learn how to save money on parts, shed weight, and improve its manufacturing methods. Even more important, teardowns help executives make long-term strategic decisions. A teardown of the 2004 Prius two years ago helped sour GM on hybrid technology. The company is slowly rolling out hybrid trucks and buses, but it's focusing its innovation efforts on fuel cells. (Emphasis mine.)
What catastrophic information did the Prius teardown reveal to the sages at GM?
A dissected hybrid lies on display in the back of the VABA, next to the remains of a Chevy Malibu. The Prius demystified comes down to this: It has 1,432 propulsion parts (the Malibu has 822). It has two electric motors, plus lots of software and finely machined gears to transport power to the wheels. That makes the Prius very expensive to design and build. (Once again, emphasis mine.)
Oh, waaaaaaa.
To Toyota, this decision to try something new in the hybrid design was nothing new. Remember, they fully expect to improve the design slowly over time, gradually improving the Prius or its successors. These first few model years are going to prove the test bed for the more fully realized hybrids of the future. That's kaizen in action. If they have to take a loss on those first few model years, so be it. At Toyota, that's how progress is made. Based upon their current market standing, I'd have to say their philosophy seems to be working.
At GM, it seems, progress has to be made in the book keeper's ledgers before it can ever leave the factory floor.
That part counting attitude shows up in the only GM hybrid design on the road (to my knowledge), the Allison Electric Drive system (Allison is a GM subsidiary that focuses primarily on truck transmissions). According to this article:
Allison Transmission's electric-vehicle think tank in northeast Indianapolis could start clearing urban smog by mid-decade with its electric solution. Replacing the hefty diesel engine with a slimmer model hooked to sizable electric motors would power the 20-ton buses just like environmentally friendly cars such as Toyota's 52-mile-per-gallon Prius. Buses equipped with Allison's diesel-electric system won't match Prius' mileage, but bus fuel economy could improve 60 percent, reducing smog and related respiratory ailments as buses burn fewer gallons of diesel oil. (I'm still doing that emphasizing.)
I drive those diesel hybrids at work. The fuel savings estimated in that article proved not so great. No one would give me specifics on the record, but the Allison system did not improve fuel economy that much. It's safe to say the number above is off the actual fuel savings percentage by an order of magnitude, at least with the buses of our fleet. I'd be very curious to learn if the newer generations of Allison drives have improved much at all.
Fuel savings be damned. Let's talk about one aspect of GM's focus that drove their research dollars in absolutely the wrong direction. Again from the Wired article, remember that little blurb about "focusing its innovation efforts on fuel cells?"
In the back of the teardown building, alongside the dissected Prius and Malibu, lie the parts of a GM demonstration vehicle powered by hydrogen fuel cells. Although the vehicle has more total parts than the Prius, almost none of them move, eliminating many of the finely machined gears and engine components of the traditional auto. The tables bearing the drive trains of the Prius and Malibu are laden with parts; those for the fuel cell car are nearly empty. The three teardowns tell the story of GM's plans: Go directly from gasoline to fuel cells with a mere nod to hybrid tech. . . .
It seems like a brilliant move. . . . But GM's critics aren't so sure, calling the company's fuel cell talk "defensive PR" because it's so far behind Toyota on hybrids. Auto industry consultant Maryann Keller points out that fuel cells are every bit as complicated as hybrids, if not more. "You're talking about a completely different propulsion system!" she says. "No one can even repair or drive one!" She argues that, instead, GM needs to follow intermediate steps. "What technology doesn't become cheaper and easier to build over time? . . . Hybrids are complex and expensive, but you'll never figure out how to make them better and cheaper until you put the technology in real-world cars." (I emphasized.)
Many may not realize the sheer ironic gravity of that last emphasized statement. Please allow me to illustrate.
Yes, GM is supposedly pursuing fuel cell technology (or was, back when they still had some money). What is a "hydrogen fuel cell powered car?" The fuel cell takes hydrogen from an onboard storage tank, combines it (usually) with atmospheric oxygen to produce electricity, and exhausts the resulting water vapor. The electricity produced by this cell drives one or more electric motors to make the car go zoom. So, to follow the sound logical argument Ms. Keller notes in her last quoted statement, GM probably should build electric cars to improve that technology. With me so far?
Alright. GM did build electrics. They produced probably the best US-built on-road electric car, the EV-1. People loved them. People had to pass a test just to lease them. Not "buy," mind you; lease. That test probed into all kinds of personal information only minutely relevant to most car buyers, asking applicants to state their incomes, verify they had a private house with a garage, and own up to all kinds of past purchases like (according to one question from the questionaire) a Betamax when they had been available. Supposedly this showed you to be an early adopter willing to take risks with a new technology. After all that, the "winning" drivers had to cough up about $3,000 to install a charger in their garage -- installing on-street chargers other cars could use were specifically not an option. Only then were they given the option of paying $450 a month for three years to drive the cars.
And once that lease expired, GM took the cars back.
No purchase option was ever made to any of the drivers, even when drivers literally offered a million dollars for as many as that would buy. (In fact, some cars were taken back early, the result of a recall. For their $450/month, those lucky few were given Geo Metros to drive through the rest of their lease. The problem prompting the recall was never fixed.)
After all that, what did GM do with its successful fleet of electrics, cars that could have yielded a goldmine of development data as it rolled through the roads of California and Arizona? 40 were donated to universities and museums, but only after the drive components were removed and the institutions signed a legally-binding promise never to get the cars running, ever again. Only one EV-1 remains intact, the one on display at the Smithsonian. The Smithsonian has a policy on authenticity and would not accept a "disabled" car . . . though the one it displays does have a dead battery pack.
And the rest? What was to be their fate? Take a peek.
They were hauled to the desert and crushed.
"Hybrids are complex and expensive, but you'll never figure out how to make them better and cheaper until you put the technology in real-world cars." Oh, and we should add, ". . . and you're willing to learn the lessons these cars can teach you."
Based on the Wired article, the movie (a story which I followed from the owners' perspective online as it was unfolding), and just about every other horror story I've heard about the GM "design" process, the company is doomed. The only kaizen they have available to them applies only to the bookkeeping department. Company designers and engineers have this Sword of Damocles called The Bottom Line dangling over every move they make, over every innovation they suggest or pursue. Yes, Rick Wagoner was asked to leave the company; but he was replaced with another executive brought up through the ranks in the same poisonous miasma of Profit Über Alles that spawned Wagoner and all his predecessors, the folks that cumulatively got GM into this escalating mess that will soon prove their demise. Oh, but don't just take my word for it. Evidence for this self-defeating culture abounds. For a look at this short-term profit culture at work from a very different angle, see bradhicks's excellent piece on the short-term gains the company got on worker pay raises verses retiree benefits that in the long run ultimately crushed all of their profits to a flat pulpy mess.
That's the insidious problem with culture, be it one shared by groups in a company or a country; one does not succeed unless one at least outwardly accepts the "way things are done around here." After a while one accepts this Way and even subconsciously absorbs it as Normal, wrong though it may be. Maybe it's in the way you develop your company's products. Maybe it's as simple as the words everyone uses so often around the office that don't exist elsewhere. Whatever it is, we human beings will overtly or subtly (depending upon the humans involved) enforce the worldview that the group uses to pursue its ends.
Worldviews don't die easily . . . at least not as easily as the companies and peoples that embrace them. Financial and physical realities have a way of pointing out the strengths, blind spots and inflexibilities in worldviews large and small. Winning cultural strategies can lead to success; losing ones lead only to death.
Addendum, Earth Day, 2009: I added the detail about the EV-1, something I had forgotten to add in the original draft.