Episode Transcript
[00:00:00] Speaker A: We have silos. We have a 12 step process. Each step in that process belongs in a different silo. Everybody performs to the beat of their own drum. So step one can be twice as fast as step two. But step one doesn't know really how fast step two is or step three. Nor do they even know what's going on at the end of the process. They just know what happening in their in their department or cell or line or whatever pole is thinking completely opposite of that. 99.9% of companies don't understand pool and they don't use pull. They use what's called Taylor's Puss.
[00:00:38] Speaker B: Welcome to why they Fail, the podcast that pulls back the curtain on why continuous improvement efforts fail. Buckle up because we're not here for motivational fluff. We're dissecting the short sighted decisions and leadership agendas that sabotage CI success.
But don't worry, we'll clue you in to the few simple keys to success to avoid these pitfalls. If you're ready for the truth, let's do this.
[00:01:11] Speaker C: Welcome to the why they Fail podcast where we strip away management fluff and examine the hard numbers behind operational excellence. I am your host, Kevin Clay, Master black belt and CEO of Six Sigma Development Solutions Incorporated. This Today we are tackling a massive operational breakdown that is costing companies millions of dollars in trapped cash. We are diving deep into the hard reality of implementing true pull system versus push operations.
Most organizations suffer from the delusion that throwing a few visual boards onto a production floor means they are running a lean culture. In reality, if your visual cards are still moving to the beat of an assumed traditional MRP lead time, you are completely trapped in a dangerous push framework that creates localized chaos and blinds your leadership team to systemic constraints.
[00:02:11] Speaker A: So Phil, how did you get into continuous improvement?
[00:02:17] Speaker D: Get hired at Toyota Motor Manufacturing Kentucky over here in Georgetown, Kentucky in 1988 and it was just right place, right time. Kevin, to be honest with you, I was young. I was hired in as a team leader, did the whole Japan thing, had a personal trainer which they call them senseis now but and I was fascinated from the start because it was so different and interesting. I learned something every day. Not that I remembered it all, but I learned something and you know, mostly by doing. Now there was some education involved, but mostly by doing. But it's just fascinating cause my first group was conveyance but they were responsible for parts delivery in the body weld and I think we had 600 employees at that time. At the very beginning how the Kanban worked and then Stamping used a different kanban. It's called a triangle kanban.
And my. I watched my trainer set everything up and it was just very. It was just very mentally stimulating, to be honest with you, Kevin. You know, if you got a curious mind, you kind of fall into that. So that's where it started. And I mean, it was. I had five different groups over 16 years, all over the body shop, including a project group. And we did things that groups couldn't handle on their own. You know, assisted with some stuff. It was just fascinating. The whole thing, whole experience was fascinating to me. Somebody asked me one time, what exactly do you do at Toyota? I said, well, the only way I can explain that is you'd have to hop in my shirt pocket for a week and follow me around, because I. I can't explain everything that I get involved in. It's just amazing.
[00:03:43] Speaker A: And you spent some time in Japan, is that correct?
[00:03:45] Speaker D: Yeah, I went there. I was H.
February 1st. I was in Japan February 1988 with 50 people I didn't know. 7 were from body. 43 men, 7 women at the time. And it was hit the ground running.
Had a personal trainer. I spent a lot of time with him, went to his home one weekend for dinner. He took me out after work occasionally. It was a great experience. He was on to be my father at that point, and he actually had a grandchild. But it was just a amazing, not only being in Japan, because one weekend we all went to Ryota on a bullet train. Well, I'd heard of the bullet train in Ryota, so it was just a fascinating experience. But I learned a lot and like I said, 95%. It was learning by doing.
[00:04:26] Speaker A: The education really got was Lean methodology what some people called Toyota production systems.
[00:04:33] Speaker D: Yes, there's a big difference now. I think initially Lean was trying to be like Toyota, but it's really blossomed out so wide open now. It includes so many other things that actually, what has happened in the last, maybe five years ago, Toyota over here in Georgetown, you know, it's the largest Toyota plant in the world. 10,000 people work over there. And they have had a partnership with University of Kentucky College of engineering since 94, called lean. Well, five years ago or about five years ago, they changed the name to True Lean and they trademarked the name. That's a pretty interesting development because I contacted somebody over there and asked, what's this? And they said, well, the Lean umbrella has gotten so big that a lot of that stuff is not part of the Toyota system. We're differentiating Ourselves, Toyota. The Toyota system is not lean and lean is not Toyota system. They're way different now. Way different.
[00:05:27] Speaker A: Lean to me is just a methodology, right? It's a, it's a combination of different tools that all, all have an output, have a, a benefit. I want to ask you a few questions.
So you know, we, we talked a little bit before we had this. You said the word Kanban. Explain what a kanban is when, when it comes to this lean methodology.
[00:05:51] Speaker D: The kanban is called it the autonomic nervous system, production system. It, it moves parts at the right time and the right quantity to the right place.
Right. There's a time element involved in any kind of kanban calculation that you do.
The kanban is a part of the pull system. It's not a pull system. It is part of the Toyota pull system. When I say there's a time aspect, if you're in a push system, you're working on lead times that are assumed to be accurate and stable and they're really neither. Because the problem with that is they cannot anticipate variability during production, quality issues, machine downtime, you know, all those types of things that happen during the day. In a pull system, Every, every process has to be able to meet tact supported, okay, including stamping, which makes things in quantity, lot sizes. But everybody has to be able to. And if you can't, that's where the problem shows up. But the kanban is just pulling because sometimes, you know, you'll run. We rent a car a minute. Sometimes we'd run for hours and never stop. Sometimes we'd stop for 30 minutes for whatever reason. The kanban stops and starts the flow of material based on the usage. So that's the magic of it. It, it keeps your inventory's levels constant.
[00:07:17] Speaker A: We'll get into a little bit more about what kanban is. I think you, you've got a lot of insight. I know that kanban is, is a small part of the pull system and the pull system and flow, they have to exist together. Flow at the, at the pace of demand through tact and pull keeps us from pushing stuff through. So I see this all the time when I go into different companies. We have silos, we have a 12 step process. Each step in that process belongs in a different silo. Everybody performs to the beat of their own drum. Step one can be twice as fast as step two. But step one doesn't know really how fast step two is or step three, nor do they even know what's going on at the end of the process. They just know what happening in their, in the department or cell or line or whatever. Then they push off and they push the not finished goods but the in process work in process. They push that off to the next station but usually it goes into a inventory location and then the, the next station grabs it whenever they have time for it. So you have this huge imbalance going on in a process. It doesn't matter if it's manufacturing, processing, a PO customer. Yeah, it's all the same thing, right?
[00:08:30] Speaker D: Exactly.
[00:08:30] Speaker A: And pull is thinking completely opposite of that. But 99.9% of companies don't understand pool and they don't use pool. They use what's called Taylor's push. You push stuff through the process, I've seen that thousands of places. What I'd like to talk to you is some of the, the things we talked about in our discussions. Moderate consultants preach that you must stabilize a chaotic process before you can implement pull. But you argue we must start pulling immediately. Sorry, let me back up and restate that. Why does the industry feel this pull first approach and what are they doing wrong from the start?
[00:09:10] Speaker D: Kevin, I wrote a book several years ago and one of the reasons was I could not understand why nobody is using pool because it just makes perfect sense. My experience, I don't see any other way to do it really, to be honest with you. But so I looked into it and this has been even since the book. And really there's two reasons people don't pull. One is they don't know how they think it's Kanban. They don't understand. You know what, Toyota has a very good pull system. And unless to me, unless you actually worked in it and was involved and then really for me I had to study on it after the fact, it would be very hard to pick it up. There's so much more involved than just a combine. But that is a problem. I call it ignorance. I don't mean that in a bad way. I just mean lack of, you know, information. The other thing is there's a heavy reliance since the 80s on MRP ERP systems to schedule production. And those are push systems, obviously.
[00:10:05] Speaker A: Well, I will tell you that MrP ERP systems, they were made in some part of having a min max, having a digital min max. But most companies don't even manage their inventory well enough to even trust that min max. So what they end up doing is because the system is hardly ever right, what you have in the system usually never represents what you have on hand. That's why you do do expedite cycle counts every day, you know, expediting and, and that ends up turning, turning what could be deemed as a good Kanban system because it gives you min and max, but nobody uses it. So therefore it ends up being a push system.
[00:10:49] Speaker D: Well, and they're set up on a lead time and that lead time inversely to a pull system. Some days at Toyota, I'm going back a year, some days, a Toyota make, let's say we were supposed to make 500, we made 4, 440, because we had a lot of downtime for some reason. So we'd work overtime and try to make it up. Well, still, we wouldn't get there. So what we do the next day, we would. We would work more over, you know, more overtime to get back to square one. And buffering, when you buffer properly, you can do that. And that's another aspect that, you know, most people would not understand if you didn't see it and be involved in it. Most people look at buffers in the obvious spots, but Toyota does it much, much differently than anybody else I've ever seen. Anyway, it's such a good system. I don't understand why people don't do it. But those are the two main reasons is one, it's just they. They've not seen it, haven't worked in it, don't know how to do it properly. Two is they rely totally on their MRP to schedule. I was at. I was at a factory years ago. They put out 30 schedules weekly, weekly schedules. So what would happen is, of course, you would be in a process, I'd be in a process. I get a schedule. You get a schedule, right. And there's supposedly, if you're downstream from me, mine is what I'm making for you. Right. Well, what would happen is I'd have the whole week's worth. So what did I do? I'd batch build everything. That made it easy on me and say, here you go, Kevin, open wide, swallow hard, here it comes, buddy. And you're like, you're looking for, you know, I gave you stuff that was due on Thursday. You're looking for something that's due on Monday for you. Well, but you got all kinds of stuff. So what do you do? You find something that you can make to stay busy, and then we've got a mess. Right? And that's what they would do every week.
It just drove me nuts. I was. I could not understand why they wouldn't, you know, think about doing something different. Here's the other thing, Kevin. You can keep Your MRP system and ERP system, you don't. There's a lot of good information in there. You just don't use them to. You don't use them to run production.
[00:12:42] Speaker A: That a good pull system doesn't need an mrp. It doesn't need a ERP system because those replenishment signals are signaled by things that you actually see, not a system that tells you in a report what's going on.
[00:12:55] Speaker D: Well, if you have a sophisticated pool system like Toyota that goes back to the vendors, you know, like they've built over many, many years, they didn't even show it to the. To their business partners for, I think it was six years when they started pulling because they weren't ready and they wanted them to see how. Exactly how it worked before they started doing that. But I'm talking about just on your. On your footprint, your local footprint. You. You can pull inside that footprint. You'll need your MRP system. For one thing. You need to know what due dates are. The MRP has due dates in it. There's a whole lot to this. Frozen schedules, all these things that matter to production. It's hard to change gears in a manufacturing plant quickly. They work best when they're stable. But they push systems just created so much chaos.
[00:13:44] Speaker A: They teach green belt and black belt, and so do my colleagues. And we have a simulation that shows a push versus a pull system. It's using dice and chips.
[00:13:54] Speaker D: I've done that, Something like that before. Yeah.
[00:13:56] Speaker A: Okay. And it's an epiphany to most companies because they all exist in a push process. So we exist by a schedule.
And then everybody produces to the beat of their own drums. Again, you got 12 steps in a process. Everybody's going at different speeds, so there's stuff building up in between. So we, we play this chip and dice game. It looks like we're playing crass, but what we're doing is we have five people in a. In a row and operator 1, 2, 3, 4, and 5. And we have the first operator roll a dice, and then whatever they roll that amount, they put that amount of chips onto the next person. We do this about 30 times.
[00:14:41] Speaker D: My goodness.
[00:14:43] Speaker A: And what happens is, is there is. There is this huge amount of variation between ships. Different people, some people might have 20, some people might have two. Right. And they keep doing maybe the wrong two. Yeah. And they only do this 30 times. And what we find is the amount of whip is pretty big. There's a lot of work in process, although there is finished goods, but the, The Proportion of whip to finished goods, it starts to encroach upon it, right? And so we then we then pull the chips back. And then we say, okay, let's do a pull process. And a pull process. We give them six ships and we say first we have to create a kanban, some kind of visual signal, right? Because they all grab the chips and put them in their hand. And I say, I can't tell how many chips you have. There's no visual signal there. It has to be something that I can see. So eventually they all put all the chips, lay them flat, one next to each other and then you can see six ships. So the, there's two columns chips and then when they start, they start from the end and then they finished goods rolls and then pulls the amount of chips that he or she rolls. And then you do that going all the way back. The funny part is when somebody has five chips and their Kanban limit is six chips. Somebody has a five and then they roll a six. They, they all of a sudden want to grab six chips from the next station. And we say that you are now, you're now in a push process because you have no limit on the amount of inventory you have. That's a. Every, every company we go into, every person we teach has really has a hard time breaking out of that. So at the end we have a discussion. What, what did you see between the two? They say, well, you know that the, the amount of inventory, the whip inventory in a push process went way up. I said, what happens if you roll, if you do this a hundred times? Well, the inventory will go up. I said, what will the inventory be in the pull system? Because there's five operators with six chip seats and that's 30 chips. And he said, if I run a hundred times, it's 30 chips. If I run a thousand times, 30 chips. So all of a sudden they see that the, the finished goods builds, right? But the inventory stay the same. But they always come back and they say, you know, but I shorted the customer. There was a lot of times where I shorted the customer. And eventually they get to understand that that shortage of the customer allows for us to see problems. Because in a push process, inventory hides problems.
[00:17:22] Speaker D: Yeah, let me say something about pull also. At Toyota, everybody is building to a buffer. It could be one, it could be six.
[00:17:31] Speaker A: And, and let me ask you this. The buffer is the, the kanban, the in process Kanban limit.
[00:17:38] Speaker D: Limit. It may only be able to hold that many, but it is the limit most of the time. There's some kind of a pokey oak situation or something.
[00:17:48] Speaker A: So that could be like a, that could be a card that has the limit on it. Like my, the max limit is this. And then we need to get down to a minute.
[00:17:56] Speaker D: It could be a lot of things depending on what you're making somebody the buffer. That's an important part of the pool system at Toyota because buffers can eliminate the effect, doesn't eliminate variability. You still got machine downtime quality problems, but it eliminates the effect of that variability on the outcome.
[00:18:16] Speaker A: Most people will listen to the word buffer and to most people, most companies, buffers just inventory. Your buffer's not just inventory. Inventory in between systems is how they, they buffer for the next next system. But it's not, it's, it's how they're just building inventory in an unbalanced system, in a balanced system where we are pulling at the end through the process and then replenishing from the end to the beginning. Your buffer has to be scientifically determined. You have to have a buffer that is the amount of time it takes you to get that product to the next step. So if you've got a very complex process and a very non complex process that your, your buffers might be different because you can produce quicker. But there has to be a limit to those buffers and that limit has to be calculated amongst the flow of the whole process, just not what you can make in your cell or your department.
[00:19:16] Speaker D: Only if you're running to attacked does that work. So buffers vary because, not just because of that. As I said at Toyota, everybody's got to support the car minute thing. Okay? So body paint, everybody's got to support it. Now some people batch build like stamping, so they, they use what's called a triangle Kanban. It's a pool system, but it pulls in batches. But the rule of thumb on a buffer is the minimal amount to keep your immediate customer continuously with work. Let's say you and I are sitting three feet from each other. Well, one between us is plenty, right? Let's say you're 20ft down the way and there's a conveyor between us or rollers or whatever. There might need to be three because of distance.
[00:20:00] Speaker A: Let, let me, let me correlate this to something that's outside of manufacturing because a lot of, a lot of our listeners are in manufacturing that. This is one thing I tell everybody that pull. It doesn't matter if you're manufacturing or processing, pull actually works in grocery markets.
So that's where it started you have canned green beans, right? And you get canned green beans. But the lead time for you to get those is once a week, every seven days. So you have to have enough canned green beans to feed your customers. You somewhat have to use a forecast for that to say, okay, I do use on average a hundred thousand of cans of green beans within seven days. Because my vendor doesn't get back to me until that seven days, I'm going to have a hundred thousand plus some kind of a service level or level that's above that based on the variation of that vendor getting back to me. So sometimes he might get back to me in nine days. Sometimes you might get back to me in six days. So I've got to have enough based on how much I can manage being out of that. You go into a grocery store and you don't see all thousand cans. All of that stuff is in their supermarkets. It's. But it's behind the scenes. And, and we know that we. In a thousand cans, in seven days, we might sell 150 cans a day. So that's how much we put out on the. On the floor. And then as they pull those cans, we see a visual signal that says, okay, we've got 12 cans. Maybe it's a POS system that said that. That we sold 12 cans. So then they go back in at night when they stock and stock back to me. That's a perfect pull system.
[00:21:43] Speaker D: Yeah. And in groceries, they have real time, they scan, they know real time. Exactly. Pretty close to exactly what they've got. The other thing is about groceries. There's some seasonality involved.
[00:21:53] Speaker A: That means that those Kanbans aren't setting and. Forget it. Those Kanbans change based on the seasonality in that business.
[00:22:01] Speaker D: For sure. Yeah.
[00:22:01] Speaker A: And that's, that's. That's. I find that to be a great example of. Of a pull process that, that I teach people that it happens. You. You deal with this every day that you go to the grocery store and it's happening while you're there.
I think it all gets back to existing to the beat of individual drums. But there is no system. Drum I in step one, don't know how fast or how slow things are getting out the end. I just get patted on the back if it. If it works and if it doesn't, I get yelled at. But I don't have any clue why I'm getting yelled at because I'm. I'm doing what I think I'm supposed to be doing.
[00:22:40] Speaker D: Yeah. And push systems just. They're just very chaotic, very chaotic.
[00:22:44] Speaker A: You, your, your data highlights that standard lean initiatives yield a paltry 11% EBITDA increase over three years when a true pulse system drives about 30 to 200%.
And this is an explosion that can happen in one year. So why is the C suite so blind to the system? And are there scorecards actively hiding that reality?
[00:23:09] Speaker D: Well, I think it goes back to two things I said earlier. It's one, reliance on the MRP system.
We'd rather chase things down and expedite and do all that business, beat up our suppliers because we're running crappy production. And the other thing is ignorance. They don't know how to do it properly. So there's fear, there's safety in numbers in this case. And so they want to make sure they got plenty of stuff so that they can make their monthly the ebitda. Let me, let me, let me get on the ebitda. Okay, so that was a study out of Europe, four year study. It was actually the first two years there was zero increase in ebitda. So you got a lean group going, a lean thing going for two, two years and you're not making any more money on it. Okay, this is a study of a hundred manufacturing plants. The third year went to 11, the fourth year it went to 19. So basically, the way I look at it, that's kind of like the, a price increase to reflect inflation. I mean, it's nothing really more than a very paltry thing. Now one of the things I think that goes on with in the lean world nowadays is they set goals in the C Suite. Well, we want 3% EBITDA. That's what we want out of our Lean initiative. That's a low number. I mean, if that's, if that's success to you, you're missing the boat. Toyota, when they started, now you think about this, Kevin, in the 50s when they started pulling, they didn't have all these things, Kanban and Hijunka and that, that was added as they went when they saw the problems happen. Tpm, tqm, all those systems came about because they would stop the line because the, you know, the buffering, all that, in spite of that and because they countermeasured it, their productivity, their productivity increased four times the first 10 years. So that would be like making 100 cars a day. And then 10 years later you're making 1600 a day. That's your productivity. Okay, now there's some increased cost with materials and so on, but you're that productive. And that was as they were learning, you know, kind of teaching themselves how to countermeasure these variability that they were experiencing. They started pulling, then they adopted these countermeasures, these systems, these processes med, all these things to keep from stopping the line. And as they did that, over time it became a, you know, pretty well oiled machine.
[00:25:34] Speaker A: You, you stated that companies fail because they treat Kanban as a standalone tool rather than an underlying system. How does this mimic a scenario where executives just play with a management fad like a shiny toy until the novelty wears off?
[00:25:51] Speaker D: Well, again, it goes back to ignorance. Kanban is a tool within Toyota's pull system. There's all kinds of tools. There's smed, there's tqm, there's tpm, there's buffering, there's tact, there's all kinds of, and they're all, they're all important. But what gets all the attention or a lot of attention is a Kanban. I've been many places in push systems and they've got, you know, we're on Kanban, we're pulling. Well, no you're not because here's why. Yeah, you're moving it, but you're moving it at the pace of your push system which is set to what it's set to lead times which are thought to be stable and accurate and they're not because they don't take into account the variability that happens throughout the day. Right. Or week or month. So you can put anything on a combine, but it doesn't mean anything other than you're just moving it. When you know, somebody sends you the Kanban back, you're going to, you're going to replenish it, that's great. But at the end you're not getting anything more out the door unless you're pulling because you're still running at the pace of the push system.
[00:26:58] Speaker A: When we talk about this concept of pull in our class, we're talking about dust and time. And we have a lot of companies, they call us, they want just in time, but they don't want anything else. They say we just want just in time. But without the infrastructure of continuous improvement, understanding lean, understanding flow, you can't have just in time. You can't just have pool. So what they want is they want to go from a just in case, you know, JIC where they hoard stuff down to adjust in time, but when they go to just in time because they don't have those tools that you talk about like smed, the different things that we have to put in place to kind of get rid of the waste in the process so it flows better, it pulls better, it just ends up going back to adjust in case. So it goes from just in time to just too late and then it goes back to just in case. That's what I created the book. Why they Fail is you can't just have pull. You can't just train a green belt. You, you have to have a structure in place. You have to have the foundation or else any, any house of continuous improvement, be it lean, be it six Sigma, theory, constraints, whatever is going to fall.
[00:28:15] Speaker C: What an incredible masterclass from Phil Ledbetter on the raw mechanics of an authentic pull framework.
This conversation hits right at the core of why so many corporate continuous improvement programs ultimately crash and burn. To recap our main insights today, remember that treating Kanban as a standalone visual tool is an executive trap. If you do not have a unified system drum, you are simply masquerading a traditional push schedule that piles up dangerous work in process inventory and actively hides deep operational defects.
We also broke down that vital European study revealing that surface level lean initiatives Yield A paltry 11% EBITDA increase over three years, which basically just matches inflation. On the flip side, implementing true pull system versus push architectures with mathematically calculated buffers can drive an immediate 30 to 200% EBITDA explosion within the very first year. I want to give a massive thank you to Phil for coming on and sharing his decades of Toyota expertise, and a huge thank you to our audience for tuning in. If you want to prevent your continuous improvement efforts from turning into another corporate clothes hanger that leadership plays with and ultimately abandons, grab a completely free copy of my book why they Fail and the Simple Key to Success by clicking the link in the show notes below.
[00:29:47] Speaker A: It.
