Lean.org - Lean Enterprise Institute | Lean Production ...



Tom Ehrenfeld:Good morning and welcome to WLEI, the LEI podcast. I'm Tom Ehrenfeld, a senior editor at LEI and I'm here with Dr. Jim Womack and we're going to discuss current events and the fact that the outbreak of the Coronavirus has left many in the media to question the viability and kind of wisdom of having just in time supply chains and how they are now threatened by events in the news. Reading these articles strongly reminded me of something that Jim wrote in 2006, and which he may have well as written this week. So Jim, welcome to the podcast.Dr. Jim Womack:Thank you Tom.Tom Ehrenfeld:Can you just start by reading from your Gemba Walks essay called Just In Time, Just In Case and Just Plain Wrong.Dr. Jim Womack:Well thanks. Sure enough, this was written in 2006 in response to the avian flu. The avian flu came after the 9/11 issue and maybe before SARS. I get a little confused about these things. Of course now we've got coronavirus. But it's interesting, the media always pretty much reacts the same way, that the world is being disrupted because we have just in time supply chains. And if we didn't, never specifying quite what we would have instead, well then everything would be fine. And because the supply chains are now spread all across the world, which lean thinking says challenge that. That's a bad tendency to spread everything out. To make each production step as far from the next production step as possible. So it means that we've got lots of stuff flowing around. But again, the attack on JIT is always that we don't have any inventories and so therefore we can't supply demand.Dr. Jim Womack:Okay. So in 2006 I wrote a little essay, Just In Time, Just In Case, Just Plain Wrong. I started by saying first, what is JIT? It's a simple idea formulated by Kiichiro Toyoda who was the founder of the Toyota motor company. His father had founded a loom company that it was based on. That was in the late 1930s. By the way, he couldn't do anything with it then, but he had the idea, and the idea was that each step in a value stream should pull precisely what it currently needs from the previous step. Pull, it's a pull system. And this would signal the previous step to immediately make new items to exactly, and this is important, exactly replace what has just been withdrawn and taken to the next step. So the idea is to replace complex scheduling systems, which depend on centralized accumulation of information and complicated formulae with simple reflexive systems. That's a key word, reflexive. Now that's another term for pull is reflexive. That work much better while dramatically reducing the amount of inventories along the extended value stream.Dr. Jim Womack:Toyota implemented its pull system by means of simple rules. One was that between every step in the value stream, it is critical to accurately calculate the standard inventory. In the early days, people were writing about zero inventory production. Toyota never said any such thing. You have to have a little bit of inventory between each step to deal with batch sizes, with shipping quantities and so forth. But the standard inventory is the amount of material that must be in place so that the downstream customer is never disappointed. And it consists, wait a minute, this standard inventory has three parts. Buffer stock, safety stock and shipping stock and buffer stock is the goods that are already finished and kept on hand. And that's to deal, and we'll get back to this in a minute, to deal with sudden spikes in demand from the downstream customer.Dr. Jim Womack:So in the case of a panic of the current sort, suddenly the medical system is calling for lots of masks. Maybe it's calling for lots of ventilators. So there's a sudden spike. So that's why you have a buffer stock. We'll get back to that in a minute. Then the second thing is the safety stock, which is finished items or raw materials that are maintained to protect the output of each step in the process if upstream suppliers suddenly failed to respond to the pull signal, right? So they could have a problem of bad quality, of broken equipment, workers on strike, whatever. So the safety stock is to protect a given step from the upstream step and the buffer stock is to protect the customer from your step.Dr. Jim Womack:Then finally the shipping stock because we can't send pieces one at a time. We all like to talk about single piece flow, but no one is going to make a surgical mask and ship them to the hospital one at a time. There's going to be a batch, so you have to build up that batch, off it goes, build up a batch. So that means you're going to have some inventory in the shipping lane, which is entirely there just to get a transport size shipment. So a second critical role, JIT is to select one point on the value stream as the pacemaker and to add additional buffer stock there to deal with normal fluctuations and consumer demand. As much as we would like consumers to behave smoothly, they don't. No where do they, there's always some fluctuation in consumer demand, and so therefore you have a buffer stock to deal with the day to day, week to week, up and down in demand so that you can run all of the upstream steps smoothly. Of course we call this [Japanese 00:05:45], right?Dr. Jim Womack:So every step upstream can operate smoothly with that, with level demand. And when done properly, leveling demand largely eliminates the need for buffer stock between each step and reduces total inventories along the value stream dramatically. That's good, by the way, that's a good thing. So what's the problem and why do commentators always suggest that JIT can't work when there's chaos in the world?Tom Ehrenfeld:And I want to remind folks that this is a piece that was written in-Dr. Jim Womack:2006.Tom Ehrenfeld:2006.Dr. Jim Womack:That's 14 years ago. Correct. Well, the problem is that severe disruptions and geopolitical, bio-geopolitical events are natural biological catastrophes, such as we're seeing with coronavirus, have to be dealt with outside the framework of JIT, only muddle thinking results when normal commerce and extreme emergencies are combined. So how should these issues be uncoupled? Well, let's look at, and I'll now modify what I wrote in 2006, let's look at the coronavirus where there's major concern about for example, surgical mask or ventilators. People who have pneumonia and coronavirus leads to a very nasty pneumonia, can't breathe. So you have to have a ventilator to help them with their breathing until their system recovers.Dr. Jim Womack:So if you really want to deal with that, governments I hate to say, but governments need to make a decision on just how many spare units completed and ready to run should be kept on hand to deal with sudden enormous surgeon demand. Now back in 2006 the U.S. government had a stockpile of about 4,500 ventilators. They really did have a stockpile, but the thought was that there would be a much larger surge in demand. It didn't happen by the way. But that stockpile, that emergency stock is simply a physical version of an insurance policy. Physical version of insurance policy, except that the policy is for society rather than for an individual. This is society's insurance.Dr. Jim Womack:Now here's what we hear instead that old fashioned, just in case inventories located along extended value streams can somehow or other solve the problem. If we just had more inventories, at sort of random points along this long stream, well then everything would be fine. We could just ramp up production and it'd be great. Got plenty of parts. But of course the problem is that there's no capacity at each step to go beyond the normal leveled load. That's why people set standard inventory to deal with standard capacity. We want the system to be running pretty close to maximum capacity.Dr. Jim Womack:So suddenly we need a whole lot more ventilators. Well we could have a kazillion parts. There's no way to assemble them. We don't have the equipment, the people, the skills or whatever to have this massive increase. So therefore it's just muddled thinking. And by the way, thinking that companies on their own, in any case could keep mountains of buffer stock or finished units is simply naive. They would go bankrupt. So what the government needs to do is to decide how to distribute the emergency stock that they have prepared before the fact that is in inventory, because in these panics when price allocation really is not acceptable to people. You can do it through price allocation and the rich people all get a ventilator and the poor people all die. That's a tough sell from an equity and justice standpoint. So that's the deal.Tom Ehrenfeld:So it sounds to me like one of the issues is that people are conflating capacity with availability. That the question of capacity relates to how much is in the system and how smoothly the system is flowing, can produce it, whereas the availability, I'm using it to refer to the amount of finished goods that are available for use and ready to be deployed or used at those emergency moments.Dr. Jim Womack:Well and it gets a little more complicated with this world that we've created where companies have tried to move the production of every step in a value stream to the lowest cost point in normal circumstances. So that labor intensive activities should be done in Burkina Faso or somewhere and highly sophisticated technological processes should be done in the U.S. or Germany or wherever. So we've created a system where there's a lot of movement. And that means you might have to have an emergency reserve at a number of points. But wait a minute though, one more thing, this is fascinating, that in the case of China, in the early weeks of the Coronavirus demand for cars fell by 90%. 90%.Tom Ehrenfeld:Consumer demand?Dr. Jim Womack:Consumer demand. They can't leave their homes. And Wu Han, nobody's buying a car, right? So wait a minute. The problem actually is a lack of demand rather than a lack of supply. And that depends on where you are. Hey Italy, you know right now, suddenly a hotspot for coronavirus. I would guess car sales aren't doing too well at the moment. So when you have a real crisis, there are lots of other things that can happen, including a collapse of demand.Dr. Jim Womack:I once worked years ago, I hate to say this, paid by the office of emergency preparedness to think about how to design a rationing system for gasoline after 1973 and the energy crisis. We had two teams, our team was trying to figure out how you would do rationing. And the second team was trying to figure out what would happen to demand. What they figured out was that if you had a real energy crisis, and by the way, '73 was just draining the system. Everybody filling up their tank and creating the appearance of a crisis when there really wasn't. But if you had a real crisis, the economy collapses. There's no demand. So it doesn't make any difference. You don't need rationing cards. You need some way to re-stimulate the economy. You're working on the wrong problem. So things are complicated, but that's sort of the big picture.Tom Ehrenfeld:Well, let's try to tease out and isolate the most important salient, lean point here. To me, this feels like, it feels like the real issue has to do with stability. That small systems and large systems are really tested in these unique moments. And one of the most important benefits of having a real robust lean system has to do with the creation of stability, and that in fact it's the opposite of what many of these articles suggest. The articles suggest that these systems are more fragile, that they're more vulnerable to spikes in demand and kind of unique a crisis. Whereas feels to me that those companies that have stable lean system, stable lean supply chains are in fact better positioned to anticipate and respond. Do you think that's a fair thing to say or you know?Dr. Jim Womack:Well, I do separate emergency from normal. We don't know when, but it turns out it looks like we have these spikes. We had Avian flu, we had SARS, we've got Coronavirus. Is this the last time that some pigs or raccoons in some country are going to slip some viruses across into the human side? Probably not. So it would be sort of naive to think that this is the last of these. I wouldn't think so, but there is a question for society which is how much insurance do you want to buy? How much emergency stock do you want to provide? That's not an easy thing to figure out by the way. What's the frequency of these spikes? How big are the spikes? What is the deterioration rate of the emergency stock that you've put aside? If it's drugs, two years and maybe the stuff's no longer any good.Dr. Jim Womack:So you'd have to have some, it's not that hard to do the math, but you just have to decide how big an inventory do we want to have and that's called that the emergency stock, to deal with the crisis. Now hang on. For the long term, when you look at the world as a commercial system, it seems to me the longterm trend is going to be that people rethink these long, long, long supply chains and do what I call lean math and figure out where the logical place to make things is. Which by the way, might include some consideration of what do you do with a big spike, right? So for example, if automobiles and their parts for the North American market were very largely made within the North American market, well then you have a very different situation then if an awful lot of the components are made on the other side of the world at a place that perhaps has been a generator of these biological crises.Dr. Jim Womack:So there are many reasons why having a supply chain extended out across the world creates vulnerabilities, but not due to Just In Time. It's simply due to the physical location and the disruptive potential that's built in by doing things in this way. Take the most extreme case. Let's suppose that, just suppose that all motor vehicles bought by Americans are 100% made in the USA. Just imagine that. If you have a Coronavirus problem in China, it has no effect whatsoever. You don't have the need for emergency stock because the system works within a bounded space. Now do I think people ought to do that? No, not by a rule, not by government regulation. But do I think companies ought to think a little bit more about their vulnerabilities? And so there's nothing wrong with JIT. You're not going to deal with vulnerability by going from JIT back to big batches that are allocated by big MRP systems that can't possibly have any effect.Tom Ehrenfeld:And I think that's the bias. I think that is what fuels the question and then into whether JIT is vulnerable, because the assumption is that JIT does away with safety stocks. Does away with the comfort zone of having excess finished goods and in parts goods. Tell us again why having an excess of say surplus parts in process is a bad thing.Dr. Jim Womack:Well, it's a bad thing because it cost a lot of money and it doesn't create the good thing, which is the ability to immediately respond to a spike in demand, because there's no capacity to translate or convert those parts into usable products. They're constrained. And hey, by the way, here's another thing you could do. Just imagine that you said, "Gosh, what we need is for every car plant that's running at reasonably full output, we should have another car plant that's mothballed.Tom Ehrenfeld:That's what?Dr. Jim Womack:Mothballed. It's just sitting there. The workers are just sitting there. This is the emergency reserve, the FEMA of automotive production. And they're just sitting there waiting for the horn to sound, which says, whoa, you need to immediately ramp up to full output to deal with this crisis. Right? So come on. No one would propose that.Tom Ehrenfeld:Is there a kind of lean, should have, in this instance, exhortation, like a lean response to something like this? If companies were to take lean seriously. If it could be deployed in a productive way.Dr. Jim Womack:Well, if you haven't got any emergency stock somewhere pretty close to the customer, you're out of luck. The lean response I advocate is people like us need to say something when particularly the gut of media ... By the way, The Wall Street Journal, which ought to know better has always been the worst on this. That when people say these dumb things, we need to say something to say, whoa, wait a minute. You've got a situation and you're proposing a solution that can't make it better. It might make it worse. So why would you do that? So lean thinkers ought to say something. Hey, we're saying something, Tom. Doing the best we can.Tom Ehrenfeld:Okay.Tom:So let me switch gears and tie it to a new LEI book, Steady Work by Karen Gaudet which is a lovely first person account of her work at Starbucks in basically building in daily stable improvable systems. And it's punctuated by the events in Newtown, Connecticut where there was a mass shooting. It was tragic. And the response at that store was essentially heroic. And it was really built on a foundation of steady work because they didn't have to think about the work, but rather could do the work, it released them to have more human interactions at a time when human interactions was essentially what they were called for. So that, to me, feels like a very notable lean take away. And I wonder if it's fair to equate that with what you're saying about the main systems and [inaudible 00:01:10].Jim:Well, it's interesting. I just saw an example from the book in real life this morning. I was coming down to LEI. I got on the Green Line, our ancient subway in Boston, always hoping for the best. I always have. Hate to make a product endorsement, but always have my Starbucks. And someone or something across the street at the office building had caused the fire alarm to go off. This was about 10:30 in the morning. And people had poured out of the building and poured straight over to the Starbucks. So the Starbucks is now full.Jim:Now that's analogous to what happened in Newtown. The people pouring into the Starbucks were the first responders who had been on... This is not just the instant of the event, but they were there for a while. And the media. So there're a kazillion media there. firemen, and policemen, and so forth who just need a break. And so they wind up with a demand that way exceeds their supply.Jim:Now what I saw at the Starbucks this morning was really very interesting. The crowd is there, and they're all lined up. And the fellow who's the team leader behind the counter stands up and announces, "Hey folks, here's what's happened. We've had this big surge. And here is our capacity. This is the limit of our capacity so that we're running now that we can flex and have more people on, less people on and so forth through time of day they do that. But right now at the minute this is our capacity.Jim:And so, therefore, oh, we're doing the best we can. We've got no wasted time. We got no wasted effort. But realistically it's going to take about 10 times as long as you would normally wait for a Starbucks. So you know, if we had massive vats of coffee on the roof, massive dark roast pipe place up there that we could be just pouring down. But our coffee system can only produce coffee at the rate it's producing coffee right now. So therefore we're doing the best we can. We're not going to just collapse because of this and the confusion and all that, which can actually happen. That output actually falls because the staff's so overwhelmed and they abandon standard work. And all kinds of chaotic things start to happen. But instead said, we're going to run at our full output, our standard pace heads down, everybody working won't have time to chat with you. But realistically it's going to take a lot longer and you might want to think about there are other places around what we do."Jim:So I thought, well that's really interesting. He knew their capacity. He knew what they could do that in trying to deal with this wave, they didn't just suddenly panic and start just doing whatever, which you've seen in lots of situations if you've lived for a long time. Okay? When things start to go South, well they actually get worse than they need to be because the system actually doesn't produce at just it's normal level. Things start to fall apart.Tom:Guy Parsons and I observed the operations at a high-end restaurant in town. And they were very good. And it was remarkably sophisticated production system to make wonderful food. Very consistent. And I talked to the sous chef about it and he said, "Yeah, Tom, we're great. All the way up to 98 and a half percent. We do everything perfectly, no flaws, great food. But then right when we tip up to the fullest, everything falls apart. And we're running around firefighting." And he wanted to find a way to make it work consistently all the way up to a hundred. Jim:And maybe you can do that. And maybe you can't. Okay? That you have to be realistic. That just, in general, when people try to go faster than they can, they discover they can't even go as fast as they could. You try to go faster than you can, you can't go as fast as you could. If you said, "This is our standard work, we can't do it faster." You can't do it. You just make mistakes. Right? so it's a sobering thing, but there are some constraints.Tom:One of my favorite essays in your book Gemba Walks is called Fewer Heroes, More Farmers... And I don't know if it's Fewer Heroes?Jim:Heroes, yeah.Tom:So where you basically saying the heroic is done on a daily humble basis? It's not creating and then heroically responding to crises that the system produces.Jim:Yeah, and there is this notion that we somehow or other perhaps we're born with, but perhaps we learn it, I don't know, in school or wherever, that if you have heroic leadership and adrenaline, you can actually produce more than you thought you could produce. This is interesting, but it's magical thinking. Okay? it's magical thinking. One of my favorite examples, of course, is Tesla, where you have the heroic leader who has the sleeping bag and sleeps at the end of the line. And is somehow or other going to encourage or exhort or threaten the lads to go faster.Jim:But you can't go faster than the equipment will permit, then the materials flow will permit, then your skills will permit. Really, it's amazing that we all have this notion that we can, if we have to, just go faster. But you forgot to pack for the trip. And the Uber is at the front door and say, "Oh, well this would normally take 10 minutes, but I'll get it done in two." Right? And then when you get to Singapore, think about all the things you don't have. So there are some constraints there.Tom:Yes, today, February 25th, the Tesla performance belies your argument. I'm not saying you're wrong. It's just interesting that they've quadrupled their stock in the past year.Jim:Yeah. Okay. But that's the stock. Tom, what's that got to do with?Tom:I'm just, yes, presenting this playing devil's advocate.Jim:Oh, okay. But look, I can't deal with that. It is interesting that they're now talking about their new Shanghai plan. And that gets us back to the coronavirus and all that. But pointing out that they've been able to ramp it up much more quickly because they're using cheap Chinese labor and much less automation. Okay? and the whole reason they got broken down at Fremont to begin with was that they were trying to heroically automate final assembly, which people have looked at for 50 years. And there's been a little bit of progress, but it turns out it's really, really hard. So the premise there was with the heroic leader, we will just find a way to do what no one else has been able to find a way to do. Interesting idea, but magical thinking. Okay? so actually Fremont got a whole lot better when they ripped out all of the automated stuff that they couldn't get to run.Jim:So that's all right. But what I'm saying is that heroic leadership... And now let's go back to the Starbucks. Suppose that you had this heroic team leader who says, "Folks, our maximum capacity would appear to be a hundred cups of coffee an hour." Or whatever.Tom:But we're going to shift into ludicrous mode.Jim:That's right. "We're going to go to ludicrous mode and we'll make 200." Well wait a minute. The coffee making equipment can't make it. The hands can't move fast enough. So forth. There are in real life, whether you're lean or you're not lean, there are constraints. And so that, actually, I think the brilliant thing about steady work is that it showed they knew what their capacity was. And in trying to surge to create so much more capacity, they were realistic and said, "We could make a little bit more up to X, but we can't make more than that. And so if we want to do more than that, then we have to get other Starbucks stores to chip in and start adding their capacity." Or whatever. Right?Tom:And so the most powerful response was based on a thorough self-assessment, an understanding of what they were able to do and to do well.Jim:That's right. And the ability, even when the pressure is on, to perform at that known, doable, maximum rate without getting distracted, without getting scared, just steady at full output. But you can't go faster than basically the equipment and the available people on hand. Okay?Tom:Fantastic. Thank you, Jim. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download