Uh so the first keynote yesterday was kill all the third leaders. Uh so I hope you won’t see me as a third leader, but at least I survived Thursday. Uh but I want essentially to share what other third leaders have done which has proven to work really well for the last 70 years and I believe it will work very well through the through the AI uh I won’t say hi, but to the through the AI time.

Um so I have quite a lot of material and I know it’s very very difficult to stay alert for one hour. Uh I’m grateful that it’s the first session of the day, so it gives you to kill your rest of the day. But I hope I can get you awake until the end of this work because the most interesting things are a bit too obvious. The less interesting is about my my past. Uh so I was really an IT nerd at the beginning of my life. I studied mathematics. Uh and then I studied IT and uh my my work for what is now called Master in IT was to disassemble an operating system called CPM running on Apple II. And it was really great to come from the machine code and and reconstruct the the code that had been written without knowing if it was right or not. But finally like 20 years later, they published the the source code of CPM. So I still have to check one by one. But anyway, this is uh I say that here because IT people normally I I hide this part of my life. And and then uh later in my career, I became CIO for DFC in France. I was also uh CIO in Paris of Europe car International, the car company. the different part of the Volkswagen Group where I was working there. And uh later I became CIO of uh logistics company called H logistics based in Paris and after that, I became CIO of Toyota Europe and finally I went to Japan. Uh and it to my big surprise because I’m not CIO anymore. I last week I got uh I got a kind of lifetime award which I I did not expect at all. And I was trying to say I’m not doing IT only and then uh and then I got this uh CIO Hall of Fame. So I had two awards at the beginning and at the end of my CIO career. Anyway, this is uh too much self-service servicing and not interesting for you. So uh I was from the last 20 years is uh from 2005 to 2020. Uh I was leading the IT of Toyota in Europe based in Brussels, which is also my hometown. Uh and then I actually left uh Toyota Europe completely and I started in Japan as a local employee in Japan for the last three and a half years of my career. Uh where I was senior advisor to to the head Chief Global strategy officer for a company called Toyota Systems Corporation in in Nagoya, Japan. So this is the grouping of the IT of Toyota Motor Corporation and three IT companies they had already. So it’s 3,000 employees and 9,000 contractors. So around 12,000 people in IT and uh and they worked essentially for Japan. So one of my roles was to to establish a global strategy for them. So you you will see them more growing outside Japan. Uh and then in 2023, I was 65, I decided to I did not decide. It was normal age to retire. Even though in Japan, they wanted to keep me 10 more years, but for different reasons including that I became grandfather in Belgium, I decided to come back to Belgium. But I was a bit bored to be retired. So I was retired for three months and then uh we created a company uh with my wife who is hiding somewhere. Yeah, in the back. Uh and the name of the company is Hakoba. So for those who know Japan and who have kid in Japan, Hakoba is a series in Japan. But if you write it with Chinese character, it is whiteboard. And those two characters is the Chinese name of my wife and it’s my Chinese name. So that’s the story of this name.

But even so from the next year, we we started to to work with a brand name which is Lean Institute Belgium because there is a global network of Lean Institute which was founded by Womack and Jones who are the first one who wrote a book about Lean uh in 1990, the machine that sells the world. And so one of them started in the in the United States, another one in in the UK, and there was Brazil and so on. And so we are number 28. There’s also IF in France that most most of you know again. Uh so we are a small sister company of IF if you want to see it like that. Uh so the company is my wife and me. And sometimes we work with a few external people. Uh but uh there’s a lot to do for two people, so we we really hope that uh Atlantic AI will do a lot of the work of people we would have uh hired otherwise. Uh and so what we do is lean executive coaching, lean training and we even sell lean books to people who don’t buy books these days. So there are few of you who buy books, please continue because not the best way to to earn money to to sell books. Yes, I say a word about our life in Japan because sometimes people are not interested in my presentation at all, but they are very interested in our life in Japan. So I always put these two these two slides at the beginning. So we were I’m lucky and unlucky at the same time that we arrived the first day of COVID. Uh we took the last plane from Europe to Japan and we arrived on the 18th of March 2020. And uh and so Japan was closed to the external world. So we had two years Japan for ourselves without any tourists. So we could see Japan without tourists and that’s it’s fabulous. So we went to uh 47 prefectures in Japan. So so we went to all of the 47 prefectures. Not during work time, sometimes but not too much, but essentially during weekends and holidays. Then from the third year, we started to to travel to Korea, Philippines, Taiwan, and so on when the roads were open again. Uh we uh we went to very interesting places like uh in Hokkaido, we could see a bear, not too close actually. We had the chance to to climb the Mount Fuji on the day I turned 65, just four days before coming back to Europe. Uh I won’t do it ever again, but I did it. Uh we had a house which looked more like an American house than a Japanese house. And uh we we did uh a hike in uh in an island in the south of Japan called Yakushima, where it goes to see a tree which is 7,000 years old. And it takes like 10 hours to go out and back to the preparation for Mount Fuji. We went to Okinawa, and we even attended the Tokyo Olympics, uh where we had a Belgian flag because no, there were no spectators. So the only thing we could see was the cycling. And as you know, there’s some Belgian people, I think in cycling, so we we were there. Okay. Uh yeah, sorry. And and on the right is the the Google Cloud of the pictures we took all over Japan.

Okay. Try not sorry. Uh yeah. So my professional life in Japan uh is uh was very interesting because at that time there were almost no foreigners. So one of the big weaknesses of Japan is they have they struggle a lot to study English. Of course, the English speakers also struggle a lot to learn Japanese, but it’s less a problem for them in the world of today. So so I gave English lessons. Uh and I also gave training to the managers which I called fit for Global to to help the managers being able to to discuss in in global meeting because Japanese people would tend to just not speak at all when they are not very comfortable, for example, with the language. And then of course, the Westerners do some wishful thinking. They think everybody agrees, but the guy who didn’t speak actually totally disagrees. Nobody notices, but this is quite a frequent situation. And uh and I had a chance to go regular. I was living in Nagoya. Uh we were living in Nagoya and I had the chance to go to Tokyo very often uh to meet the people in Tokyo at the corona open by Toyota. Those who on one hand are building the woven city uh on the Mount Fuji. Uh which is now first phase is finished, so now there are really people living there. So it’s really worth going and and seeing what Toyota has done there. It’s kind of lab for mobility. Uh and also in that company, there were half of the people hired from all over the world. So it was a very interesting cultural laboratory to see the interaction between these Japanese guys and the guys from all over the world. And they also developed an operating system for cars which is called Arene. Which uh Toyota will sell to other car makers as well. Uh and then I had roles for Toyota Motor Corporation. In terms of course, since I was the only foreigner there, they were saying, oh Willem, you do this, wouldn’t you do that? So so I had uh nice full days especially with the meetings around 11:00 p.m. with Australia to North America. Uh and I was global sponsor for Toyota production system and uh and for agile practices. And I I did something that uh practically I I saw very few people who did. I gave TPS training to Toyota in Japan. Because they they had not practiced uh TPS everywhere in the company and in particular in IT, they were very reliant on external suppliers who who did not do agile or TPS. And uh so because I had practiced uh uh agile with a flavor of TPS in Europe, I I found myself in this position to to teach them, which is very difficult. It’s not so easy for a Japanese person of course to accept that uh a guy in Europe, a foreigner will tell them how to do TPS. Uh but I learned a lot by doing that. Uh and then I was sponsor for some other stuff. So the incentive of what I found is uh when I started to do agile uh with my team in Belgium, the European team, nobody questioned why should we do agile? Everybody said, oh, something new, let’s do it. But in Japan that’s not how it works. They they asked the famous five ways. So they will ask uh why do you think it’s better? And then you really have to have a much stronger uh argumentation to tell them why it will be better. And while doing this, at some point, you can question yourself, is it really better? So so there are questions like oh, the requirements of my users won’t change in the next five years. Why should I do agile? So of course, I say it’s impossible. I’ve never seen any users who don’t change requirements in five years. But it’s okay, in Japan, you see many many interesting things. There are things they have Kaizen to death for many many years and they they felt they were like perfect and there was no no break in the next five years to to change whatever. So Agile has left for white in that case. Or uh okay, what if the migration is only technical? And I had to say, yes, but the technology also changes. So if you you do your your migration in waterfall over five years, but the database you use is DP2, and when you have finished, nobody uses DP2. It’s not bad to be Agile even for technology. Anyway, we had many interesting discussions. The most crazy one that I had not anticipated is uh there’s an agency in Japan called IPA. It’s not a beer. It’s the is the IT something agency. And uh they are actually telling the companies how they should work with IT. But because it’s a state organization, they don’t react very quickly to the new trends. So they had not adicted uh rules on how to do agile well. And so people come back to me. They said, sorry, we cannot do agile because uh because the IPA doesn’t tell us how to do agile. Uh I say yeah, but we are Toyota, the biggest company in Japan, everybody respects Toyota, why don’t we start? They say, yeah, but this is an IT problem. Maybe we should ask NTT or Fujitsu, a big Japanese company to start with that, and then we will follow. Of course, it’s a very ideal way to lose a few months. So so I was very angry and we had many fights about that. But finally, the IPA did something, also put them a bit under pressure, and and we could start. And uh and also the natural question is, why you don’t just use Toyota production system for IT? And that actually more or less what we did, but we we can use the Agile wording uh and or we can use up Toyota, the TPS wording. So instead of talking about the stand up meeting in the morning, we can call it Akai, which means morning meeting and which is used all the time in the production of Toyota. So you have to translate a bit the concept and then then you realize that the the manufacturing people are actually pretty much on board because that’s what they wanted to do for a long time.

Uh Okay. So in the end, we had in that company, IT company, we had around 1,000 Agile practitioners. Uh I wanted to reach 1,000, but uh uh but I saw yeah, it for some reasons it it was anyway, push is not GPS. So I didn’t push too much to have everybody on board. Uh and I think now still not the case. to talk about the stand up meeting in the morning, which I call it Asaichi, which means morning meeting and which is used all the time in the production of Toyota. So you have to translate a bit the concept and then then you realize that the the manufacturing people are actually pretty much on board because that’s what they wanted to do for a long time.

Okay. So in the end we had in that company, IT company, we had around 1,000 agile practitioners, three years. And I wanted to reach 1,000, but I thought, yeah, if for some reason it was anyway, culture is not TPS, so I didn’t push too much to have everybody on board. And I think now it’s still not the case.

Uh, so, if we look at the traditional TPS house, which I guess most of you have seen. But then I will still ask why you haven’t seen the pillar called Jidoka, because without that one, the the temple is really not holding very well. so we really have those. Why do we have these two columns? historically first, the Jidoka is an invention from the first job of Toyota, which was to make automatic looms, the Meti automatic. and there, the the basic idea for quality came. And then the just in time came with the founder, Saichi Toyota and then his son, Kichiro did just in time because Japan after the war, they had no no money, they had no space for stock. So they needed to find another way to make a car with less stock and also pull flow, or try to satisfy the customers one by one.

instead of producing a lot of stuff to put on a big pile as it was very easy to do in the states. Um, and so I I will not explain everything, you will see why there are hundreds of concepts in lean, but I will really focus on the Jidoka part today. but when I tried to, I I did a PhD in Strasbourg when I was 56 to 59 years old. And in in that PhD I tried to map all the the lean and TPS concepts that I was learning in the factories to the temples, and it became very crowded because… First of all, every consultant has another version of the temple, because of course they have to have their own version, and and the place where they put things is sometimes on the basis, sometimes in the middle, sometimes in the top and there’s no real consistency. So what what I did is this, which of course you you will all read in detail. so I took all the concepts I learned and I I grouped them with top concepts that will be a bit more readable, and to to follow up on the talk on philosophy we had yesterday. So this is called an ontology, I call it the ontology of Lean. and uh, it it shows some people can call it a mind map. It shows the relationship between the the various concepts. So by the way, it’s freely available because of my PhD is is.

can be found on tesis.zh. You will see the link at the end. It will become a book that will be a bit easier to to read. That’s my target for this year. So the top concept I have come up with are we you see that just in time and Jidoka are still on the two sides to remind people that they were the pillars.

And then at the bottom, you have the fundamentals, safety first, continuous improvement, Kaizen, who is the visualization and Monozukuri, which means to make things, so the the craft of making things well. But in order to make things well, you first need to make people who are able to make things well. So and it can be products or or services. So at the top line, you have the the people related boxes. You have customer first. And on the right you have Hitozukuri, so all the all the techniques to put the people first and to to train to let the people grow so that they can prepare the good products or services. And in the top center you have stakeholder satisfaction.

Uh, which is very important because we don’t work just to to have a big profit, but we want to be sustainable over a very long period of time, saying like 100 years. So if I if I compare with Agile, there are a number of things which you can find back in Agile. Customer…

Uh, customer first, yes, if you have a good product owner, somebody who really knows the job of the customer and has a clear vision of what the product should be, maybe you can put the customer first. But in many cases you get a product owner who doesn’t know the work very well, maybe that was the person that was available at the business to support you, and not necessarily the one who knows the job best, and then you can still in Agile deliver the wrong product, that is. Agile is more just in time.

Because you deliver smaller chunks of things quicker. and Kaizen, yes, but in Agile the Kaizen is in a very small increment, but there’s something very important is the direction, the compass, which is called Hoshin Kanri. and if you don’t have the Hoshin Kanri, you can have agile teams that go really in all directions. And never achieve anything because they don’t have the longer term vision, because they thought that was not needed anymore. So I discussed with Jeff Sutherland, who founded Scrum, who founded.

He said, of course, you need that, but because he doesn’t say it in the in his document, then some people have stopped doing it.

Uh, and then, yes, visualization, of course, you have Kanban boards, you have Jira and so on. But sometimes we still prefer a good old Obeya, which is a room where at a glance you can see the status of everything. On paper, of course, COVID was terrible for the Obeya. people could not convene anymore. So a lot of things were done with IT, but maybe it’s a challenge for IT people that nobody has been able to do something with IT that is up to date and that looks like a paper Obeya. So there’s still there’s a company called IObeya, who is trying to do exactly that. But for me, it’s not yet not yet completely there. Um,

And then I’ll come to the things which are more positive. So, Agile doesn’t always satisfy all the stakeholders. There’s not much in Agile about you should support your community or things like that, or your country or the world. And making people, yeah, there is some development of people, maybe the scrum master will be will try to develop people, but it it’s not as systematic as what you find in in Toyota. and then finally, there’s nothing about safety first.

There’s nothing about safety first. But it’s on environment, safety is really first and if you don’t do that, it’s super dangerous.

And on the right, the the Jidoka, which I call sometimes stopping time. But you will see why. there’s very little about about this aspect. There’s one thing maybe if you do a sprint and you have a impediment escalation.

Uh, that can be compared to to an Andon in in Jidoka. And so the focus today is is the Jidoka. and of course I can talk for a good days about all the other ones. So, if I zoom on the ontology, it’s still not readable probably. Uh,

but I will go, I will explain essentially three methods of of Jidoka.

Uh, and uh, first of all, what is Jidoka? What does it mean? People have invented words for this in English. some people call it automation, but because the word automation doesn’t exist in English, that’s not more clearer than saying Jidoka.

Uh, all they say automation with a human touch. And but that’s too long. So that’s why most people call it Jidoka.

And for those for those who read the Chinese characters, Toyota has even modified the Chinese character to to have this. So Jidoka normally is not written with the small, it’s actually a person.

Character which is in the red elite, normally you don’t need that. So when you type it on a keyboard, you won’t get that. Toyota has added that, and that’s precisely the human touch. So they have put a human on the left of this Chinese character. and what does it mean? Is that, you want a machine that does things as automatically as possible, but you also want the machine to realize when there’s a problem and stop by itself. And the the origin of this.

Uh, that could be a decision.

So.

So what you have on the picture is the the museum in the house of the founder of the dynasty, Saichi Toyota, and his mother was working there at night on on this weaving machine. And sometimes the thread was breaking and so she had to stop her work, repair the thread and then continue her work. And her son saw that this was really a terrible job and there must be a way to help his mother. And and what he constructed, it took him many years because he finished only in 1924, is the type G loom that you can see in museums in Nagoya, which stops automatically when the thread is broken. And and this is really key for two things. The first thing is an idea to help people and you will see that the whole idea of Jidoka is to help people. So that’s the deep meaning of Jidoka. And then also it it couples the men and the machine.

Because if your machine stops when there’s a problem, you can easily have 30 machines and each machine will say, I have a problem, I have a problem, you won’t be you won’t have to look at the machine all the time at one machine. So it also multiplies productivity by because you have this automatic device. I never found when his mother died, so she never enjoyed this invention, but because it’s much later than the start of his activity, but but that’s the intention, to help people. And so many people say, Jidoka means to stop the line, like stop the line in the in the car factory. Yes, but that’s just a visible thing. Why do you stop the line? You stop the line because the worker has a problem, he doesn’t know how to solve it and he asks for help. So, so this is the key thing. And and this thing in your organizations which are very diverse, what would it mean for you to enable everybody to stop the line and ask for help. Typically, you have new IT people starting in the company, they don’t know anything and they are struggling, and when they ask their colleagues, their colleagues are all very busy delivering something for a customer and nobody is available for help. In a Toyota factory, there must be somebody available because they stop the factory and if nobody comes, the factory stops, it’s not very good. So it’s really a way to force people to help people. And it seems counterintuitive because stopping a whole factory is a huge cost, but it really helps focus people on what is really important, and of course, once the problem is solved, it doesn’t happen anymore.

So in IT, in IT engineering, because Jidoka is not only for a for a factory, it’s also the title of my talk today to talk a bit also in in the engineering of products. of course, there are some differences. in engineering and in production, essentially in production, you repeat the same task many times, sometimes thousands of times per day, whereas if you design a new car, maybe you have a task that you do only once every time you design a new car, so to save one second is not necessarily very significant. But in the end, all the principles of TPS apply.

But you need everyone needs to know their own job to see how to apply them in an intelligent way. And that’s my passion now is to apply TPS to all activities of human, sometimes very difficult, we can apply very well to government if they want, we can apply to agriculture, we we can really apply to anything.

So I want to talk now about the three methods of Jidoka.

You know Andon, if you’re French you say Andon. No, not so many.

Yeah. Okay, if I wait long enough there will be more. Ehm, you know Poka-Yoke? Okay, quite a few hands.

And possibly you don’t know Jikotei Kanketsu?

Uh, okay. A few, few people. Maybe the reason why people don’t know Jikotei Kanketsu is it’s a bit more difficult to pronounce.

Uh, but it is super important. I would not hesitate to say it’s really one of the secrets of Toyota. Because there’s something that is difficult to do, that people don’t like to do, that is boring to do, so there’s no consultant that spend time to to explain it to people. So please bear with me and I hope you will see why I think it’s very important.

And also very much against what human beings do naturally. So but I will go quickly to Andon and Poka-Yoke and focus on Jidoka. few few people. Maybe a reason why people don’t know Jidoka and Ketsu is a little more difficult to pronounce.

Uh, but it is super important. I would not hesitate to say that it’s really one of the secrets of Toyota. Because there’s something that is difficult to do, that people don’t like to do, that is boring to do. So there’s no consultant that spend time to to explain it to people. So please bear with me and I hope you will see why I think it’s very important.

And uh it’s also very much against what human beings do naturally.

So we always go quickly to Andon and Poka-yoke and focus on Jidoka.

Andon. Andon is actually a lamp. An is to go and don is the lamp. It became a lamp that you have in the in the production. This was an ancient paper lantern. but sometimes people talk about Andon boards which can show colors like a green means you’re on track for your production and red means you’re behind.

Uh, and it’s also the Andon cord. So this is the cord that the worker will pull in a factory to stop the line. Now, people have an image that the worker stopped the line and thousands of people stopped working, it’s not like that. The line is stopped and and the supervisor comes and try to solve the problem within the cycle time, which means that the rest of the line will not have to stop. But if the problem is not solved within the cycle time, then the next one stops and gradually everybody stops. So it really happens that way. there are even today in a Toyota factory, there are 2,000 Andon calls per day. sometimes it’s a very small thing, but you have new workers all the time who need support. Even if a worker wants to go to the toilet, he will pull the Andon, the supervisor comes and does the work for the few minutes he’s at the toilet. So, so this is also an Andon that will likely never reduce. So the target is not to go to zero Andon.

And there are anyway always new problems.

Uh, that is very interesting and I spent a lot of time to think about Andon in IT. Because if you’re a young programmer, it’s probably a bad example, a young person using AI to program something. you you have a problem. Are you going to stop the the 8,000 other developers who work in your company just because you have a problem? So that seems a bit too too exaggerated. But what Toyota did for the factory also seemed very exaggerated.

So maybe we should have somebody bold enough to apply it that uh more than I did myself in my IT. but if you’re in a in a sprint, uh and you have maybe six people in the team, uh then it it really makes sense to stop and have the five other people helping the one who is stuck. So that is what could happen at the cycle at the cycle time, uh will not impact the other team. But then if you have to to consolidate what different scrum team have done, uh maybe you need also to to pull the Andon and have people think that oh, this module and this module don’t work well together. So we need an extra sprint to make them work together. So this could be also this could be a good decision.

Uh, and uh for yeah, impediment escalation to top management. When when people started to do Agile in my teams, I said, why I thought you would come to me every day with impediments because there are things that don’t work that you cannot solve.

Why you never come to me? So something doesn’t work if people uh let the their arms fall and say, oh yeah, that’s anyway what they never told me before. So we’ll not even try to solve it now. So you really have to encourage people to to escalate impediment to top management. The good thing for that is the Obeya, which means the big room where where you put the status of all projects and where the boss comes not not to say, hey, you have a problem situation but project, what did you do? But you say, oh, there I see you have a problem, how can I help?

So yeah, that’s what I said just now.

Uh, so the Poka-yoke. Poka-yoke is actually better than Andon because Andon is you have a problem and you say you have a problem. But Poka-yoke is you make sure you don’t have a problem. So normally I should start with Poka-yoke. If you have many good Poka-yokes, you should have less Andon. Poka-yoke means error protection. In the past, they said Bakayoke, Baka means fool. but for different political reasons, this word changed and we still say foolproof in English. so it’s a foolproof device to prevent, for example, if the worker picks the wrong part for the car that comes, you hear a beep beep and he knows, oh, he has picked the wrong one. I have to pick here. So there can be 1,000 Poka-yokes in a factory.

Uh, we work also in healthcare and I’m always scared that you go to a hospital, there are maybe two or three Poka-yokes to prevent mistakes. But in a Toyota factory, you have thousands of Poka-yokes. But in a hospital, I can lose my life. So I really encourage them to have more Poka-yokes to prevent this.

Uh, and the yeah, I think yeah, for for IT you have a checking for bank accounts. So you should type the wrong one it will tell you.

Uh, at least every 97 error maybe still the wrong one, but most of the cases it will detect it. Or display your address on the map and then you think you’re in a city but you’re in the middle of nowhere. You can immediately see that the address is wrong. There are many such examples.

In daily life, normally I do it interactively, but it’d be difficult in this in this room. to with a gasoline station, what is it?

So yeah, I I think you said the right thing, but I’m not sure I heard. But you maybe you cannot put uh diesel diesel in your in your gasoline car. But unfortunately, you can put your gasoline in a diesel car, which means that this is only a half Poka-yoke. It will help you one thousand of two. So be careful if you have a diesel car, it doesn’t work.

Uh, or things like uh yeah, the the safety belt. Uh okay, if you start without your safety belt, the car will do beep beep, it’s not a real Poka-yoke if your car starts. Your car doesn’t start, it’s more of a Poka-yoke. Then of course, in France, people tell me, oh, it’s easy, you put it and you sit on it. Yeah, so there’s always there’s always a common that I know.

Uh, or ATM is now they they notice that if they give you back your card before they give the money, it’s more likely that you will not forget your card. If if they give the money, you will go with the money and forget your card, which will be swallowed by the machine. You also Poka-yoke. And or in the in the metro, there are some, I’m not sure about Paris, but some metros where you put your card, you can forget your card in the machine and then you go through on the other side. That’s not your your ticket. Some metros force you to take your your ticket in order to get it, otherwise it doesn’t open the door. But if you think about it, you can have Poka-yoke everywhere in your life. And that’s what I showed with a cushion on the right, you may wonder what that is. it’s just that we have an alarm system in our home. In the if in the evening, we put the alarm system downstairs but not upstairs. So, if we are a bit tired in the morning, we go down the stairs and the alarm starts. So we just put a put a cushion at the top of the stairs and when we see the cushion, we know we have to disable the alarm. Just to say you you can really apply it in in your daily life.

And some some Poka-yokes are good, some are not so good. so we go a bit quickly on that, but these are exercises I do. So, a typical thing with a coffee machine which has cups. If you’re unlucky, you go and the previous one took the last cup and you have no cup. It will take you some time to get your cup. So, so I ask people what what would you do to avoid this situation? It’s interesting to learn Poka-yoke design because there are people who say, oh, just put somebody next to the machine who will always refill. Yeah, but that is a very expensive Poka-yoke.

Uh, and we we had other I don’t have the perfect one yet from all the answers I’ve just have in many places in the world. But uh an idea was to have a a red cup a bit before the end and the person who gets the red cup is the one who is obliged to refill. And if not, everybody will see who has the red cup. Maybe it’s a bit too much finger pointing, said some detail. Anyway, so think carefully about your design and don’t hesitate to iterate and to find better Poka-yoke that that will help you a lot. and then a typical thing in IT is you know, you always have the disk that are 80% full and then it tells you. And then nobody cares because there are thousands other messages for the operators and then a few days later, it is 100% full and the system stops. I’m sure nobody had that in their career. And and sure there, if you want a real Poka-yoke, you you need sometimes you need to give uh more rights to the, for example, the operators who are there at night. They may see this, but they don’t have the authority to to increase the size of the disk or to force other people to do something about it. So, you really need to design ways to to do it to to really force people to do something because it is a serious problem.

This is uh a bit of examples for for IT.

So we don’t always think it’s a Poka-yoke, but just the the mandatory iteration uh indentation, sorry, in the Python programming or in uh you there was a list with all kinds of parentheses. If your editor of program forces you to indent, then you will see that you have forgotten some things visually. now of course, there are things like linting, which uh prevent errors directly at coding time. So, they will tell you, not good, not good at the time you write the code. So, of course, that’s cheaper to solve the problem at that time.

Uh, and for cyber security risk, there are immediate notification of SQL injection, which can be very dangerous later. Instead of having people checking later what uh uh and being angry at people because they didn’t look at this SQL injection and now we had a data breach in production. so the next one I said already. Um, users of IT systems. I I said already some examples. Yeah.

Okay. Now, this is the most technical slide I have and I’m talking about things I have never done myself partially. but really the language you choose for programming can be better or worse, can be can have more Poka-yoke embedded in them. One very famous thing is the the new safe languages because uh null pointers create a lot of issues. So there are languages that don’t allow you to have null pointers, which seems it’s the case of Kotlin, Rust and Swift. and uh some language implemented as an opt-in, which is not ideal, they have to do for compatibility issues, but uh that means they don’t force you. So it’s not a complete Poka-yoke.

Then the functional languages, you can result what it is. That exists, there’s a language called Lean. Did anybody program in Lean in the room? Okay, I need to talk to you because it’s funny that the language is called Lean and I don’t think it’s a link with him, I don’t know. So you will tell me. interest and test driven programming. So again, if you know the concept of Poka-yoke, it’s easier to recognize a Poka-yoke or to recognize the absence of a Poka-yoke and make the right choice for tools.

Then Jidoka, Jidoka and Ketsu. Good, I’m there before the end of the hour because it’s my social.

Uh, so it means Ji if your fare, that’s the first character. Kote means process and Ketsu means complete. So it says what it says. So it means that you are able to complete the process by yourself.

So in English, they found the name built-in quality with ownership. Of course, if people don’t care about the quality of their job, it’s unlikely they will do a good job by themselves, but that’s not the most frequent case. Most frequent case is people don’t know what is a good job.

So when the head of quality of Toyota reflected deeply about why there were so many quality errors. He thought about the grocer at the corner of the street. He said, why, why did the grocer usually smile? He smiles because he has contact, direct contact with his customers, and when the customers think the products are good, they tell him. So he gets good feedback that motivates him. So when there’s a problem, the customer will also tell him and he will have an opportunity to to solve the problem quicker because he gets the feedback not indirectly through many other people. And that’s really the basic philosophy behind Jidoka and Ketsu.

So, employees and surely in big companies is very often not the case, employees need to understand the value of their work. they need to have a sense of responsibility and pride of their job. How can you, if you do something you don’t know who will do something with it or not, it’s difficult to feel the pride and the contribution to the company. Immediate feedback fosters quicker improvement, positive feedback increases self-confidence and work becomes more interesting due to this.

So the typical way to do quality assurance in the past was you do the work, you deliver the car and in the end the quality department comes and says, this is very bad. It’s not first of all, you have many people in the quality department and their job is to tell others that their job is bad. So their job is a shit and the job of the others as well. So, so the logic of Jidoka here is to really embed the quality into each process and be able to judge the quality of each process.

So, this that’s why I say it seems easy in theory, but it’s very difficult to do in practice. feedback increase, their confidence, and work becomes more interesting due to this. So, the typical way to do quality assurance in the past was you do the work, you deliver the the car, and in the end the quality department comes and says, this is very bad. It’s not, first of all, you have many people in the quality department and their job is to tell others that their job is bad. So their job is, shit, and the job of the others is, So, the logic of JKK here is really embed the quality into each process and be able to judge the quality of each process. This and that’s why I say it, seems easy in theory, but it’s very difficult to do in practice. We we are only two in my company and we try to do it with each other and of course, still many times I ask my wife to do something and she has understood something different and then we we fight. And if you want to do it very well, it takes time. You don’t always want to take this time and so on. So basically, you have necessary conditions and judgment criteria. The necessary condition is you have a good design of your process product, you have a good process and you have a capable operator. If you ask somebody to do something they are not trained to do, very likely the result will be bad. the judgment criteria is also very interesting because is the person really capable to check by themselves that the quality is good? If that’s not possible, it won’t work. It won’t be good quality. And more importantly, the authority to stop. This is why I have Andon twice here if the operator is not allowed to stop when there is a problem, of course, there will be problems because they will know their quality is bad but they are not allowed to stop, so they will give the bad quality to the next, for example, because there’s a deadline. so, so authority to stop and the skill to judge. If they don’t have the skill to judge the quality, they don’t know what good looks like, so they will just give to the next one and the next one won’t be satisfied. If you have established all of that, your internal or external customer will be satisfied. And you will always have a feedback loop with the next time. again, not easy to do. If some of you start doing it from tomorrow, and it works, please tell me because I don’t have so many companies who are doing it well. Hopefully this tutorial also, but So, there is, steps to establish JKK work processes and if you know the Toyota way problem solving, it also has eight steps and here it’s the same, the steps one to three, four, five, is actually just planning. Six is the actual do, seven is the check and eight is the is the act. So it’s a PDCA cycle where the P planning is the first question. So, you first clarify the purpose and the target. If people don’t know the purpose of their job, they are not motivated to do their job. you clearly visualize the final output. So how does it look like for you? And then that’s what you should strive to get with quality. Then you write down in detail the process and procedure and then you define those judgment criteria and those necessary conditions that I showed. and then you accomplish the work and of course you will find things that you can improve and that’s step seven, you reflect on the work and you pass on the knowledge gained to other people who can also improve their work.

Many people, especially in the West, find this boring to do, so they don’t do and they don’t have quality. So, I think we have to discipline ourselves to do it. And there are five levels of maturity in JKK. You always start by having quality gates because the quality is not good yet. You try to see which type of problems you have. The quality gate stops and checks, so there you’re still in an external quality phase. Then each uh quality problem you find, you feedback uh to to the place where the problem occurred and you do basically a Toyota way problem solving. So you solve them these problems one by one. That’s number three, and only then you establish necessary conditions and judgment criteria. Don’t try to do everything at once because we tried many times, it doesn’t work. You really have to do it step by step. So, and I don’t like to hear only Japanese people can do that because it’s never true. I I’ve seen people in the factories doing it from any nationality. But yeah, it’s hard work, but when you have done it, then we are back to what we said yesterday. You’re happy because the work flows, yeah? And and your colleague doesn’t give you shit because they have also done this and you don’t give back to the next because you have done it. So this is really what we should all try to uh, try to have. I I will skip a bit but this this is just an example, a simplified IT example, where a person is developing a program and giving it to to an operator, and, uh, you look at at the steps. And so the person who writes the program observes the work of the customer, prepares requirements, develops the program, tests the program, demos the program. I know now we do many things DevOps and so on. But what I’ve seen many times in my career is you know the operator have to run programs and they and they were complaining because, because the developers didn’t know what is the work of an operator and so they didn’t design programs for being run smoothly. And when when there was a problem at 2:00 in the morning, nobody could help. The documentation was not good. Again, you you need to to check that there is good design, but in design for good operations. The best way is to send the developers to do operations and back to developers. Those who have done that understood much better. there must be gone to go criteria, as we call, in Japanese, based on actual problems that occurred in the past, what to do about them. and of course, operator should be capable, should understand what he’s doing, what he’s playing in trust and eliminate the defect of beauty, which, is called Poka Yoke for those who have learn about the lean manifesto.

Uh, and then, yeah, you you have judgment criteria. So, good operator love to receive this program because of usability, of course, does the developer design, I don’t put it in production, by himself maybe not because the deadline is today, the the user wants it to this, so he knows he doesn’t have the documentation but let’s put it in production.

Uh, and the skill to judge. Can a developer judge what is a good program for operation? Does the programmer even know?

And then you do PDCA, you get the, very often when a program is in production,

the developer does something else, doesn’t get so much of the program in production. How to organize the feedback loop.

Uh, okay, so I know this is this is not you do workshops on that. There are those small things. So, in general, and that’s huge, but there are there are really very big benefits of doing JKK. It’s an investment up front, that’s always a difficult part. But it helps you eliminate local optimization, create opportunity for managers to check progress, deepen communications across divisions, maximize the unique strength of each partner, in increasing information sharing, to be applied to to get the feedback to discuss, reduce meetings, reveal unreasonableness to improve, improve, improve that. Reduce failures and non conformities, increase productivity and increase motivation. So of course, everyone says, why don’t we do it, yeah? Because it’s tough, it’s tough and you need to put work in it and it’s always easier to find a short cut, which seems good for today, but will be a problem tomorrow. Of course, your own DevOps is, is an example of just in time and Jidoka tension. because you the death tries to be just in time, but when you put in operation, sometimes you have a push back because, it’s not it’s not safe or it doesn’t comply with some standards and so on. and so you need to check that every change is a good change. Kaizen means change good, yeah? So Kaizen is not just a change, it’s a change for something better. So that you have to check everything.

Now, I just have two slides about the info IT engineering specifically.

Uh, so I said already, to reduce five seconds of operation to four. When you have it a thousand times per day, that makes sense in production. In engineering, okay, if you have a new car concept every three years, reduced five seconds for one of operation doesn’t make sense. It doesn’t mean the the principle of Lean don’t apply, it just means you need to apply them in an intelligent way. And then for another talk, there there are many things I can say about set based concurrent engineering. the knowledge database that Toyota keeps for 50 years, even if it’s just on paper, at least you didn’t have 20 generation of system that was the data in the meantime. And I will say in French, for those who speak French, you may know the book called the Toyota Engineering Story, which was written by the French chief engineer of the car that was made between Peugeot, now Stellantis and Toyota. So the 108 of Peugeot and the C1 of Citroën and Toyota was the Aigo. So all three cars were made in the same place. And so there’s a French chief engineer who worked with the Japanese.

And I find it really interesting that this guy, Olivier Stolier, he he is employed by Stellantis, but when you read the book, you want to work for Toyota because he’s totally enamored of Toyota in the book. And it’s it’s full of anecdotes and it’s a really easy to to read so I recommend if you’re interested in engineering to buy this book. You can buy it for me but more logically here you can buy it from SDC, you can also. And then there are books about, Toyota project development system, Morgan and Laker is a very famous part. But wrote by people who have never done engineering at Toyota, they studied a lot but they they were not at Toyota. So they still more to that than what is there. So lean product process and so.

And then I have to say a final word about the chief engineer.

Uh, because the notion of chief engineer at Toyota, is a mystery for many people. Because, when Toyota will design a new car, they always appoint a chief engineer. The chief engineer has a complete responsibility about the car, it’s a it’s a profit and loss responsibility. But the chief engineer doesn’t have all the people who work on the car under him or her, because they are female chief engineers at Toyota. And, uh, to become a chief engineer is really like you’re a semi-god, yeah? you must be fully respected in the organization that the organization will entrust you this role of chief engineer. So that means the people who do that they they really have an authority, that a natural authority. But they use it very carefully. So they very rarely over over run the decisions that the big phylos are doing, like the body shop, the paint shop, the assembly and so on. but it happens very rarely that they do. When they do, people respect their decision. So I imagine in some cultures is very difficult to implement that.

Uh, but it works, yeah? and that’s also why not all Toyota cars are the same. for example, Shiko Kaku, who is the the chief female chief engineer who did the Toyota Yaris, she was the first lady chief engineer.

Uh, she she designed a bit more space for the back for for a woman and things like that, so that makes it a car which is more suitable for for women. So it’s really the personal touch of the chief engineer. and some prefered uh uh prefer to have a perfect audio system, then they will say a little bit more money on something else, but the result of the car and the sales, the whole life cycle of the car will be attached to that chief engineer. They come to the motor show to present, to do marketing and sales of the car, but they have also done all the engineering before. So it’s really an amazing job. One of the most famous one is who did the previous. Uh when he heard he would be chief engineer, he told his wife, she said, okay, I will bring a bed to to work and I will stay here for the for the next three years, at least, but he became chairman of Toyota afterwards. and if you think about IT, what I’ve seen very often is when you need a product owner, first of all, the product owner is not always considered a semi-god.

Uh, and you may really get the person who has time, who is normally not the person who knows the process best and who can design the best project. So please start right for getting the staff here and don’t hesitate to make them staff. But staff with humility, which is the chief engineer of Toyota doesn’t show too much arrogance. So that also very difficult job they do.

And so, in summary, we want to create pull flow from the customer need. So that’s the very well known, just in time, pillar, pull flow of TPS. And the Jidoka does the country, the Jidoka stops the flow. And this is really this apparent contradiction that will make you more and more successful because every time you stop the flow is to improve something. So by stopping the flow, you make sure that you will stop the flow less and less in the future. And of course, I have to say something about the AI, this a device. Please think deeply about these concepts and use the AI when it makes sense. You can use the AI to help you design a good for Poka Yoke, why not? But don’t forget to have Poka Yoke, to have Andon, to have.

That’s it. thank you, for that I’m able to say this. this report is for my PhD, it’s 250 pages in English and a summary in French at the end. but yeah, I’m trying to put this in a more easily consumable form, but you have there also the complete ontology. and and you can reach at the Institute whenever you want. Thank you.