Hacker News new | past | comments | ask | show | jobs | submit login
The Toyota Production System (2013) (toyota.co.uk)
224 points by louis-paul 5 months ago | hide | past | favorite | 114 comments

Closely related to the need to walk the Genba, this key principle suggests that to truly understand a situation you need to visit in person. The Toyota Production System requires a high level of management presence on the factory floor

From personal observation this is very true, Japan has a great and unpretentious grey collar culture. There's a way in which all work is seen as equally essential and managers aren't above anyone else. Not just at Toyota, but even waiters in restaurants are treated like this, I think it shows in how negatively tipping is seen in Japan.

It's also a culture that exists here in Germany in Mittelstand companies, although sadly less and less in large firms where management culture from the Anglosphere seems to have gained popularity over the last 20-30 years. American management always reminds me of the historical British army, where the gentry directly was recruited into the higher ranks American management and workers seem equally stratified.

>It's also a culture that exists here in Germany in Mittelstand companies

As a tech worker in Germany I've worked in 2 different established Mittlestand companies(not startups) that are pretty much world leaders in their niche and I promised myself I would never do that again.

Their company culture was terrible and while they kept preaching things like "we're like a family here and we're all equal" that was just a facade. Nepotism was rampant in both cases, with most employees being either related or close friends of the CEO and upper management, both coming from old-money families, generating a strong tribal behavior, so anyone who wasn't your boss could get you in trouble if they decided they don't like you and raises and promotions were allocated based purely on the tribe's perception of you and performance judged on butt time in your chair rather than results. As a foreign tech worker there who's not part of the tribe you're treated with an air of "you should just be grateful we gave you a job" even though they kept whining about not being able to attract new talent and grow. The lack of transparency was prominent with the CEO always preaching how "times are tight and we're all in this together" to justify lack of wage increases while he bought more vacation homes abroad and exotic cars. Information about the company's projects, customers, sales, revenue was all kept away from the employees who were not part of the tribal circles as it only circulated via spoken words.

Now I only work for big corps and it's been a huge improvement, with American corps in Germany treating you the best. They're transparent about their projects, customers and earnings in their quarterly town meetings, information is communicated electronically, not by speak, leadership and and workers are (mostly) hired and promoted based on merit and transparent goals and processes vs straight up nepotism and tribalism.

This is exact my experience working for German Mittelstand. There are patriarch, family and their slaves. As a foreigner one ends up in the slave group. While cutting your salary because of difficult economic situation patriarch orders a new Mercedes. That was so surreal.

I wasn’t that lucky with American company. I ended up in one with American brand on the building and rotten hostile culture inside. Local patriarch is hostile towards Americans and the family insults new American owners on every occasion. Luckily I am leaving this show in 7 months for my own startup experiment.

You simply worked for a shitty German company and a great American one.

I worked for quite a few young(!) small- to medium-sized tech companies in Germany, and most were like GP describes - unpretentions, genuinely down-to-earth management.

Of course, the opposite also exists, especially with old-school managers in well-established companies. Family-owned businesses are often the worst. The butt-in-seat mentality is very prevalent. During the Corona pandemic, they had to be dragged into the 20th century, kicking and screaming.

I don't think generalizations can be made in either direction.

>You simply worked for a shitty German company and a great American one.

I worked for two shitty German companies that, ironically, are well reviewed on Kununu and well respected in the tech field (hardware engineering). I also interviewed at many more established Mittlestand tech(hardware) companies and I saw the same shitty patterns(butts in seats culture, micromanagement, no work from home allowed, outdated tech, poor pay, "flat culture and family atmosphere") so I feel safe to generalize at this point.

>I worked for quite a few young(!) small- to medium-sized tech companies in Germany

Well the ones I worked at were part of established hardware tech Mittlestand that keeps getting positively exemplified as why Germany is so successful vs other countries, not the hip new software startup.

>Of course, the opposite also exists, especially with old-school managers in well-established companies. Family-owned businesses are often the worst. The butt-in-seat mentality is very prevalent. During the Corona pandemic, they had to be dragged into the 20th century, kicking and screaming.

Well, yes, that's exactly my experience. The experience made me give up for good on ever working for the Mittlestand business, especially the family owned ones.

Out of curiosity, in which city are you located and wich field of tech are you in? I'm open for moving once covid blows over and scouting for spots already.

As someone working in Japan, I can agree to:

>Japan has a great and unpretentious grey collar culture.

but I disagree on:

>There's a way in which all work is seen as equally essential and managers aren't above anyone else.

The latter I've seen to not hold true at all, and this assessment can be derived with equal weight from the very "top down, respect your seniors" culture of Japan that you're taught as early as elementary school over and over again.

I've seen this go in to effect so bad that clueless managers that are by no means an expert, overrides expert assessment, and people including the expert reluctantly budge to the pressure of "your boss/manager is always right".

The more realistic take would be "all work is seen as inherently valuable and should be taken seriously" to be the general mantra in the Japanese society, often associated with ikigai [0].

This societal pressure of "finding ikigai in your work, and strive for perfection" combined by "respect(unconditionally) the decision of the elder/boss" is probably an contributing factor to why Japans "Job satisfaction" level is around 42% [1]

[0] https://bit.ly/3585zVr

[1] https://www.weforum.org/agenda/2020/06/job-satisfaction-glob...

Quote from [1]: "Only 42 percent of Japanese said they were satisfied with their work and, to add insult to injury, 21 percent said they were dissatisfied, both the lowest and the highest outcomes in the survey, respectively."

Your ikigai reference [0] link is broken (403 forbidden)

The taxi drivers in Japan take their profession extremely seriously and they are very good at their job..

Interestingly, I’ve heard one profession that is not very well respected in Japan is software.

I’ve heard this as well. Both from westerners who worked in software in Japan and from Japanese people (all men) who worked in software in Japan. I also work with someone who worked in design in Japan for 7 years, who has been a great insight into the working culture there.

Can anyone elaborate or clarify why software developers might not be valued in Japan? I recall one person on HN explaining it as a problem of the market being saturated and software not being as huge of a market as it is in the USA. I can’t speak to that, though.

Ironically, based on the original comment, the issue is that East Asian countries (like Germany, which was the archetype industrialisation model for East Asia) are largely run economically for the benefit of large corporations. So developers are hired en masse like every other employee (in particular, skilled labour is hired once after graduation), there is a stigma against switching jobs (even if you could find somewhere to move), employees are essentially interchangeable, and there is almost no innovation.

The original comment is hard to understand. Germany and Japan are two of the absolute worst places on earth to be a consumer/worker in the developed world (the IMF even produced a paper on this in an attempt to put pressure on Germany to improve standards for workers). The US is nowhere close to having a management aristocracy because the system is open. Even in the UK, which has a similar economic model, the low-level of competence among managers is actually a major factor behind weak productivity growth. People just long for systems that aren't like the one they have, low levels of employee turnover are not a panacea, they led to heavily rigid management and are often linked to a high proportion of unproductive, family-owned business that are managed by birthright...professional management is, innately, open. If you look at Denmark, high levels of employee turnover actually lead to higher wages because it creates so much more innovation (most countries that have any other model, including Japan, are trying desperately to get away from it).

Afaik, France is the only country that has the aristocracy model: it is usually literally impossible to be promoted into management if you didn't go to grandes ecoles, and most of the top jobs in govt and business are just rotated through this group of people. Unlike the US, the curriculum of related courses often has no relation to reality, and is designed to be as complex/abtruse as possible so entry to the higher reaches of society can be limited. It is difficult to understate how envied the US system of management is compared to this...it isn't just that the US does this slightly better, it is that the US is at 100, and everyone else is at 10. It isn't close.

> So developers are hired en masse like every other employee (in particular, skilled labour is hired once after graduation), there is a stigma against switching jobs (even if you could find somewhere to move), employees are essentially interchangeable, and there is almost no innovation.

I'm not sure why you're getting downvoted. I worked with German multinationals for about a decade, and I saw quite a bit of this type of behavior from their management.

Japan is often said that here's best place to live but worst place to work. This is reasonable because harder work makes inexpensive and useful services (example: restaurant). Is Germany worse for both?

I’m curious, which part of 36hrs/wk work, a legal cap at 40hrs/wk, 6 wks of vacation, 2 yrs of parental leave, socialized health care, college (essentially) for free, etc, is the most horrendous part for you about being a worker in Germany? Which makes it “the absolute worst place on earth“?

The legal cap is 48hrs/week, max 10hrs/d. Parental leave is 12-14 months and you get around 67% of your average net income. When you look at vacation you have to take public holidays into account and that differs a lot between the states within Germany. Germany has 5 public holidays (which don't fall on a Saturday or Sunday), the USA for example has 9 public holidays this year. The benefits are largely the result of workers unions. Most benefits were given to prevent a rise of communism. The result were high work standards. But those high standards only apply to some industries which are in decline and some of them have almost disappeared (steel for example). Germany is almost certainly not the worst place to be a worker. The problem is that Germany is a gerontocracy. Only the elderly part of the workforce fully benefits from them. 30 days paid leave with 38-39hrs/week were normal a few decades ago. Nowadays one should not expect to get more than the bare minimum. In a lot of industries workarounds have become the norm. For example workers in slaughterhouses are hired as contractors so work standards don't need to be applied. The main issue I have is that the German economy and the idea of the social welfare state is on shaky feet. The baby boomer will retire within the next decade but the German welfare state is based on a pyramid scheme which will collapse soon. So Germany is a bad place to be worker in that sense that you need to take a gigantic bet on the continuation of a system which is designated to fail on the long run.

I agree with all of this, and certainly with your correction of my sloppy wrong comments (wrong numbers).

Further, what you sketch out in the second part is exactly my biggest squabble with Germany. Let's see how a society that lives off of exporting machines fares in the future after fighting innovation and entrepreneurship for two generations now.

That said, I still don't see even in the slightest how these are "the worst working conditions on earth".

> correction of my sloppy wrong comments (wrong numbers).

It was not completely wrong. Being an employee gives you those benefits (eg you can stretch parental leave to 2yrs with part time work, the minimum vacation is 24 days instead of 30, but there's unlimited sick leave), and it is incomparable to the ridiculous PTO/PSL system in the US.

However, companies by and large only do this due to external pressure (govt, unions, competition), not any kind of inherent cultural benevolence.

In tech/engineering, this typically means that instead of direct hiring, hiring through agencies where people are regularly rotated out of projects to not have to become permanent employees, and without the regular benefits of the host company. Freelance tech work is actually becoming rare because German law has Duck Typing for determining employment status, so if you're de facto salaried, and the govt finds out, both the employer and employee suddenly owe all the relevant employment taxes, even years back, and because this has been abused and has been cracked down on, nowadays getting freelance work is difficult.

Of course, that somehow didn't stop companies from trying to pull the same scam with "gig economy" BS on the lowest paid workers the past few years.

You are missing out one relevant point: currency.

It is easy to win at exporting if you pay employees in a functionally worthless currency. Germany's trade surplus is 10% of GDP, this is roughly where China was at peak. Germany can win at exporting because: consumers are conservative and will accept large corporations milking them, and other countries in Europe are willing to let Germany export deflation to them.

It is one of the most unsustainable economic models (and the source of significant global financial instability...China's trade surplus peaked in 2008, not a coincidence) and that is before you consider the need going forward for high levels of innovation, business creation, etc.

I didn't say on earth: I managed that section, I said in the developed world. If you provide a counter-example then I will accept you are correct.

I believe the same is pretty much true for The Netherlands, which might not be a surprise.

The pensions in The Netherlands might actually be better arranged for now, but it's likely at some point that the EU will (indirectly) try to gain control over Dutch pension funds (e.g. by forcing these funds to allocate a certain percentage of spending to buy Italian bonds).

Health care isn't socialised, it is a mixed-system (almost nowhere has a "socialised" system...even the UK, the most effective systems are mixed like the US system...Germany, in particular, is known for having fly-by-night private doctors who will do all kinds of crazy shit if you give them enough money, it is one of the only places in Europe where you can get stem-cell therapy for example...it is a very odd system). I am not sure what you mean by "legal" cap, if you want to work more or less hours you can. Other similar countries do not employ slave labour. And education is very cheap in the US (cheaper than most countries in Europe) if you go to a state college...at private universities, the fees in Germany are still very expensive (the US has better colleges, so private colleges in the US are always going to be higher than everywhere else)...you aren't comparing like to like.

And presumably, you don't come from Germany because all the reasons why it is bad are reasons that wouldn't occur to most people who don't live in that system: unions that are essentially non-functional/run by companies, staggeringly low levels of consumption relative to income (Germany runs a trade surplus worth 10% of GDP, which is all funded by savings...this is as high as China when China was at peak inequality), the average wealth of German citizens is equal to Greece...an economy that isn't functioning and doesn't have the large number of billionaires that Germany has (who largely got rich in the 30s), non-existent levels of competition, non-existent levels of innovation, closed banking system that pays nothing to savers and funnels free money to bankers/large corporations who have (historically) lost it all in short order, sometimes shockingly low wages (I believe this has changed but German apprentince wages were sometimes 2 or 3 euros/hour). Again though, read about German history (or even Northern European countries, generally bar the UK)...the economic system was designed by large corporations, that is who it is for. It isn't for consumers (or google "IMF Germany report"...if you don't believe me).

> Health care isn't socialised, it is a mixed-system (almost nowhere has a "socialised" system...even the UK,

> the most effective systems are mixed like the US system...Germany, in particular, is known for having fly-by-night

> private doctors who will do all kinds of crazy shit if you give them enough money

I think that's missing the point a little bit - when we talk about the UK and Germany having "socialised" healthcare this usually refers to the fact that it's not primarily provided by a mishmash of private insurers and private hospitals run as businesses for profit, but by the state as a service for the populace. It doesn't mean that there's not also a small private healthcare market for the wealthy to get quicker treatment in fancier facilities or to get unusual or elective procedures. One side effect of this is that there’s basically no anxiety about losing your access to healthcare if you lose your job or choose to quit - it’s just not part of the equation. This is the main point being made here. Oh and I think we should be very careful using the words "effective" and "US system" so close together in a sentence when talking about healthcare :-)

There are interesting examples of countries who have "mixed" systems though - Netherlands (where the healthcare company I work for are based) and Czech Republic (where I live) both have a sort of hybrid system of a market of health insurance companies in tandem with govt funding. And while ČR's isn't particularly spectacular, NL is very highly regarded worldwide. Frankly I'm a little surprised that those who are against M4A or single-payer don't ever cite it. Could be because the government still contribute more than the private insurers.

I have no idea what you mean...Germany is the European version of the US healthcare system. It is very high-cost, the private sector is not small at all, and it isn't only used by the wealthy. Again, German doctors are kind of notorious.

The reason why people don't mention NL is because it doesn't fit anyone's narrative. In the US, they have a mixed public/private system with a large private sector so their solution is: someone else should pay for my healthcare. Mixed systems are less attractive because, to a large extent, that is what the US has already (the US is exceptional because of the administration/ancillary costs in their system, not the structure). I agree though, NL is very good. The results are pretty average but it is very well optimised overall. I am from the UK, and healthcare is unbelivably bad because it is so ideological...that is definitely the wrong road.

that's one way to paint a picture - with facts, actually. However, as parent said: There's another picture. Mandated 20 days/year holidays (in big companies it's often 30 days), overtime has to be paid (these are things that came to pass due to pressure from unions!), healthcare for everyone, rent prices that are not horrendous (and a lot of social housing), a taxation system that is friendly to people with low income, unemployment benefits, minimum wage (9€/h) etc. There's a reason so many people look up to Germany, even in Europe. Your opinion is definitely more fringe. btw. I'm German.

>overtime has to be paid (these are things that came to pass due to pressure from unions!)

Not all jobs in Germany are unionized and many modern contracts have optional overtime hours built in them to prevent the employer from having to pay up for crunchtime.

I did not define what I meant by socialized health care, but I meant it in a positive way. Everyone is insured, and your insurance is not bound to any employer. Ironically, this means there is some actual competition in Germany (or at least more so than in the US where you can choose between the cheap and the expensive option your employer signed up for.)

No, you cannot work as an employee for as long as you want. My numbers were wrong, but it's a 10hr cap per day, and 48hr cap per week, as written in another comment. In its default, beyond that, your boss is personally responsible for this.

I am sure all you write about he financial system is true, but the thing is that compared to the US you can get away with much less savings and liquidity, and you get a much higher standard of living for a lot less money. Compare rent, cell phone, internet, health insurance, and you get a lot more for a lot less in Germany. With your health insurance not tied to an employer, you also don't run the risk of losing it when losing your job - e.g. in case of severe sickness, when you need insurance most. You don't need to save up for college of your kids, and can just send them to a public school.

[edit: typos and added schools}

36h/week in not the norm in Germany. Most my friends working there have 40h contracts and some have optional overtime included in the contract to cover crunch time.

So your anecdotal evidence is better than mine? ;-)

IMO most worker is not valued well and just IT engineer in non-IT corp isn't an exception.

I've picked this up through consulting work. There are several large Japanese companies that located to this area about 30 years ago, and a few Japanese suppliers have sprung up around them. In dealing with them on side jobs and through friends, it's become very apparent that they are FAR less willing to spend money on IT than American companies. (One supplier, doing many millions of dollars a year in sales, is still running their company on Excel spreadsheets.) Which is a shame, because, from what I've seen, they're so far behind, they're smack-dab in the middle of the biggest-bang-for-buck area of the IT spending curve.

This little abstract documentary is about a Japanese taxi driver / photographer. Really cool:


> American management always reminds me of the historical British army, where the gentry directly was recruited into the higher ranks American management and workers seem equally stratified.

As a veteran, American management seemed to take straight from the American military management. I always assumed it was an artifact of WW-II management theories and the subsequent post-war academic funding boom.

Nope. American management was created in the late 19th century with railroads.

In fact, management didn't exist before the US invented it. Most large companies in the UK were run by people who had no interest in management. They bought a factory then promised a piece rate to a labour gangmaster (this is an abtruse point but wage labour wasn't really a thing until the 20th century, the model was proto-industrial "business" structures which paid by output...wages shift risk to employers, and so that took time to develop). In Germany and much of Northern Europe, there was a very well-defined capitalist and worker class who negotiated through govt (because the capitalists mostly owned the govt and got their money to build stuff from state-owned banks, most unions ended up being allied to those capitalists/govt because revolution was taken quite seriously...left-wing unions didn't really exist). Again, there was no real concept of professional management...you negotiated with workers en masse, there was total seperation (and there is still is in Northern Europe, to a large extent...most billionaires in Germany got rich in the 30s, for reasons best forgotten, and have just passed it down unchanged which is why wealth inequality is so high).

So it was with the railroads that actual professional managers started appearing (railroads were undoubtedly a new kind of complexity but US ones were run far more professionally), and shortly after this that you see management training and the like. The US was, and still is to a large extent, one of the only places where you see universities actually relate to the real world i.e. you have universities teaching practical tasks like management (I am in the UK, management is still thought of as a "joke" subject by most academics...stuff like Classics and Politics is far more respectable than business, people who need to work for money are looked down upon as hopelessly gauche...it is kind of funny that people believe the opposite is true, but that isn't self-evident if you come from a society where practical knowledge is valued).

Btw, the defining book on this is Alfred Chandler's Visible Hand. He is probably the greatest historian of business (yep, he is from the US which is the only place where historians actually study this kind of thing en masse).

>most billionaires in Germany got rich in the 30s, for reasons best forgotten, and have just passed it down unchanged which is why wealth inequality is so high).

This is the truth so I don't get why you're bering downvoted.

I remember seeing some study that showed much less social mobility in Europe compared to the USA. Many families in Europe that were wealthy hundreds of years ago are still wealthy to this day, whereas in the USA one (used to?) see much more transition of wealth.

Perhaps in this sense, Europe is much more unequal than the USA.

Bloomberg wrote about this issue: https://www.bloomberg.com/news/articles/2016-08-23/how-to-st...

As did Famcap: https://www.famcap.com/2016/05/2016-5-17-wealthspike-europe-...

Yes, there are families in Florence, England and other European countries and cities who have enjoyed unbroken dynastic wealth over several hundreds of years. But this doesn't prove that Europe is more or less equal than the US. The US has only existed for ~2 centuries. In 1870 the US surpassed the UK to become the world's largest economy, and remains so. So there's only been 2 centuries or less for dynastic wealth to form in the US, as opposed to ~10 centuries in Europe.

Yep, that is correct...although something of an inevitably given the different historical paths (and isn't true of all countries within Europe either). But what I would add is that downward mobility is just as important as upward mobility: the US is, I think, unique in that rich people can move downwards which doesn't happen in other places.

Btw, neither approach is more optimal than the other. There are trade-offs. What some countries in Europe get wrong is that they are trading off innovation which is going to become more important for growth going forward.

>the US is, I think, unique in that rich people can move downwards which doesn't happen in other places.

Which IMHO is very healthy since it creates space for the younger players to move up vs having a reign of feudalistic families owning most vital sectors of the economy for decades or even hundreds of years.

The entire concept of social mobility being easier in the US is false, and has been for decades. The idea that downward mobility is unique to the US is also distinctly false - not only does it exist in other countries, it is more common in many places than it is in the US. https://www.weforum.org/agenda/2020/01/these-are-the-10-coun...


Thanks for the insightful and well written comment, however I wonder if (and propose that) both can be true: your statements regarding the initial seed of business management from western railroads and also post WWII military influence on mid 1900s business management practices.

No, the influence was actually the reverse. Very briefly: railroads, oil/telecom companies, car/retail companies then WW2 then govt begins to try and pull management from companies into the military.

This is the military-industrial complex of Eisenhower. Companies were integrated into the military during WW2, and their influence was maintained after: management practices from companies into the military. Kennedy's govt was full of people from business making foreign policy/military decisions (with unfortunate consequences). Again, it is difficult to understate how far ahead the US was with management training so it is natural that this innovation spread into the military.

This makes sense and my original comment reflects my ignorance of pre-WWII history.

> In fact, management didn't exist before the US invented it.

Wow. Just wow.

> I think it shows in how negatively tipping is seen in Japan

To be fair, obligatory tipping is pretty rare outside of America.

Not-so-coincidentally, so is the capitalism-uber-alles political landscape that requires it for service jobs to be in any way sustainable.

American unions encourage that in a number of ways. I don't know why unions don't like good people going into management but they do

I've oft wondered to what extent MBA programs, and more generally the notion of a managerial class originate from British Aristocracy.

To the contrary, the managerial class of hired administrators who did not themselves own the business was a product of the industrial revolution. MBA programs sprung up because the huge new corporations (steel, oil, etc) needed a steady supply of people to fill this role.

At least that's what I remember from Rakesh Khurana's book, From Higher Aims to Hired Hands: The Social Transformation of American Business Schools and the Unfulfilled Promise of Management as a Profession.

As a huge Ed Deming fan, I'd recommend these books:

- The Essential Deming: Leadership Principles from the Father of Quality: https://www.amazon.com/Essential-Deming-Leadership-Principle...

- The New Economics for Industry, Government, Education: https://www.amazon.com/Economics-Industry-Government-Educati...

- The Toyota Way: 14 Management Principles from the World's Greatest Manufacturer: https://www.amazon.com/Toyota-Way-Management-Principles-Manu...

Deming is really a great thinking in the area of product development and agile methodologies.

Another great perspective is This American Life's episode on the NUMMI plant[1]. Coming full circle from Deming back to GM.

[1] https://www.thisamericanlife.org/561/nummi-2015

Seconded. It's such a good portrait. The way the workers talked about the change sticks with me years later.

Adding to the recommendations, I found Rother's Toyota Kata very valuable in understanding how to make Lean approaches work: https://en.wikipedia.org/wiki/Toyota_Kata

For applying it in software, I loved Mary Poppendieck's books, which are listed about half-way down the page here: http://www.poppendieck.com/

And for the math-inclined, Reinertsen's "Principles of Product Development Flow" is very insightful: https://www.amazon.com/Principles-Product-Development-Flow-G...

Thanks for Poppendieck and Reinertsen links.

As a fan of Drucker, Deming, Goldblatt... Today's "agile" has always confused me; I really don't understand when and why our profession jumped the shark.

I was part of the early "agile" community back before that name existed. Occasionally I'll talk with somebody from back then and ask them how they think things turned out. The general answer is, "Well this isn't what we intended, so I've moved on to other things." Which is basically my answer too.

My best guess as to how the shark-jumping happened: http://agilefocus.com/2011/02/21/agiles-second-chasm-and-how...

The TL;DR version is that once it went mainstream, Agile was defined by the mainstream, which was not looking for excellence. Aided vigorously, of course, by the certification scam that is Scrum, where meaningless credentials were profitably bestowed by people who had little incentive to tighten things up. And one thing I've realize in ensuing years is that a key element was the culture of managerialism, which is inimical to anything that might make a manager look bad or diminish his empire.

re managerialism: I used to complain that we geeks got no respect. What I'd give to be ignored and neglected again.

I'm going to chew on your "second chasm" notion. Thanks.

Maybe it's just how things go. Fads, "selling out", poseurs, fashion, lost in translation, etc. Uncle Bob also references the rapid growth of our profession, where warm bodies are added faster than best practices can percolate.

To do list item: Maybe Everett's Diffusion of Innovations talks about this.

I've been getting a lot out of David Graeber's books, most recently The Democracy Project and Utopia of Rules. They're the first description of workplace democracy (collaboration, empowerment) which matches my own experiences. Back then, I was just kinda winging it (eg "What would Drucker do?"), because I really didn't have a lot other options.

Graeber considers the paradoxes better than most sages. eg How a movement begats its own destruction.

I keep thinking about that cliche of how anything taken to its logical extreme becomes it's own opposite, related to an abundance in one area causes a deficit in another.

Bringing us back to Goldblatt's Theory of Constraints. All balance is completely lost with all the players trying to hyper optimize their own little corner. Worse, "rationalists" weaponize fallacies like "beware the slippery slope!" to actively reject any kind of nuance, moderation, judgement, balance.

Thanks for listening. Trying to articulate my grievance helps me organize my thoughts.


Oh. One parting thought. I'm trying find a rhetorical basis for advocating moderation, proportional solutions. Something akin to rational altruism. I want a mashup of algorithms, game theory, and hedges (eg basket of investments, NPV) to guide decision making and governance. To make it okay to do 100 crazy ideas, because maybe 3 will hit the jackpot. To make it okay to try a variety of mitigations, because there is no one right-sized solution. Etc.

Ah, I haven't had a chance to read much Graeber. Although I appreciated his notion of bullshit jobs; it sums up a lot. And good luck with your quest! Feel free to drop me a line if you get anywhere.

I also highly recommend listening to this entertaining episode.

To answer what someone else asked in the thread, the reason other plants trying to copy NUMMI failed is that the GM managers could only copy the things they saw of the Toyota system. The lines on the floor, the equipment, etc. They did not and could not (in time) copy the mentality and attitude of the workers and management, to believe that quality was paramount and that stopping the line was ok in the name of improvement.

The attitude sank everything, and is hugely revealing. All the best tech in the world will not save you from the wrong attitude.

I haven't listened to the podcast yet, so - why NUMMI we're not able to copy the Toyota production system - given that Toyota has sent them experienced advisors ?

NUMMI was a success, but that success was not replicated at other GM plants. NUMMI closed during the recession as GM discontinued brands like Pontiac due to severely reduced demand.

Today a large part of the land is a Tesla factory.

Yeah, they cover it pretty well in the podcast. By the numbers NUMMI did well however to achieve those figures they upended many traditional roles and couldn't get the leadership in GM to get on board.

There's a great little footnote at the end of the podcast about a small upstart company buying the plant with the vision of building electric cars in the future.

They did actually. I can't remember why specifically they closed the plant in the end. But one of the overriding laws of global capitalism is that it's possible to do everything right but and still lose cause of a trade deal or an acquisition or a global recession or...

GM pulled out. GM already pulled out of that factory in 1982 and left the workers with nothing.

By the time Toyota came back with a JV agreement w/GM in 1984, the NUMMI factory was the only Toyota factory with unionized labour force. UAW Local 2244 embraced Toyota’s methods and processes. But when GM quit(again), Toyota had no need to be there anymore with a local unionized work force. How would that jive their non unionized labour force everywhere else?

GM quit. That’s what they did. They were quitters. As was Ford(quit in neighboring Milpitas a few years before GM). Toyota was brilliant. They did good to the workers they let go when they quit NUMMI. They did good by almost $280 million. Towards the end, NUMMI only rolled out Toyotas.

It’s all Tesla now. For a while Jobs ran his NEXT manufacturing from NUMMI facility before he was welcomed back to Apple. That place is legit magick.

By all accounts, NUMMI was a great place to work and with GM gone, Toyota did not need the UAW Local 2244. Why would they? It is also California’s insane regulations and strangling policies that drove the last auto manufacturer out. Until Tesla came into the picture.

And CA almost drove Tesla away recently during the covid year..who knows what the future holds.

> Deming is really a great thinking in the area of product development and agile methodologies.

“Agile methodologies” are basically Deming/Lean approaches reinvented with more reliance on subjectivity and just-so stories and less attention on coherent theory, measurement, statistical analysis, engineering rigor, etc.

It's interesting to read what other field's version of agile is and what we can learn from it and improve our own processes. Another over I enjoy is John Boyd who in essence created agile for killing people.

The Agile Manifesto does bring some Lean ideas to software development. But "agile methodologies" are by themselves a subversion of the ideas on the Manifesto, and the most deployed ones are completely antagonistic to Lean.

> But “agile methodologies” are by themselves a subversion of the ideas on the Manifesto

I think that varies quite a bit from methodology to methodology; certainly, the ideas of the Manifesto operate a higher level than methodologies, so the idea that a canned methodology is a solution rather than a starting point is contrary to the Manifesto, but aside from the potential that the methodology will be treated that way many of them are in other respects consistent with and actively supportive of the ideals of the manifesto.

Deming's 'The New Economics' is a great book. Just ignore the part about intrinsic and extrinsic motivation. The rest is solid gold.

I highly recommend the books "Toyota Production System" and "The Machine That Changed The World" for a background on the manufacturing cousin/ancestor of agile software development.

I wish there were newer studies though that cover the rise of Hyundai/Kia quality in the last two decades. If you know of any please let me know.

“The Machine that Changed the World” is an excellent book detailing differences between the Japanese style and American style of designing and building cars.

Near the end of the book is one of the most important parts, yet nobody seems to talk about it. Workers and suppliers were committed to the system because the automaker was committed to the worker and the supplier. This is the cornerstone to the entire thing working. You can’t do the Toyota Production System without it, and few companies are willing to do so.

Absolutely mind blowing to watch that transformation. And not just in durability and features but in aesthetics as well. A BMW interior is still a BMW interior but the interior of a 2020 Hyundai is much closer to the interior of a 2020 BMW than the interior of a 2002 Hyundai was to the interior of a 2020 BMW.

This is one are where others impress as well. Even GM and Ford have come far in this regard.

Guessing that last 2020 was supposed to be a 2002?

Recently read "The machine that changed the world", and I too wish there was a more recent study. I've worked in several Michigan and Indiana tier 1s and OEM plants, and seen some of the differences described, but the book is 30 years old now. The fundamentals are still fundamental, and culture change is really, really slow, but everyone's making improvements.

> I wish there were newer studies though that cover the rise of Hyundai/Kia quality in the last two decades.

I realize this is a tangent, but the fact that Hyundai managed to build this and sell it with a 10 year warranty is incredible.


Not only that! The story behind their newly founded N division is very interesting. They poached a long-time BMW executive Albert Biermann who headed multiple vehicle developments and the BMW M group, and tasked him with bringing up the N high performance divison within Hyundai (N as in Namyang R&D, but he says also Nurburgring). Since then he's been promoted to President & Head of R&D. As my most recent comment history says, I've spent several weeks in Namyang and I've never heard of an executive's name being (positively) invoked so many times inside the garage :)


There are a lot of amazing concepts that Toyota developed. It requires a lot of discipline, focus and commitment to implement properly.

One thing I didn’t see mentioned that Tesla is currently doing is a very strong relationship between design and production. The upcoming Cybertruck for example doesn’t have paint and it has an exoskeleton design that should be much easier to produce. Over the years I have not seen any radical designs from Toyota that suggest that the ease of production was influencing radical design choices.

Designing for production became highly unfashionable because the big three did it extensively in the 1970s and 1980s resulting in very efficient production of cars nobody wanted.

It's one thing to try something new. It's another to re-try something that failed (or was correlated with failure) within the lifetimes of people still in the industry.

Toyota doesn't buck trends like that. They're very conservative.

Tesla on the other hand takes risks.

Were the cars not substantially less expensive to produce? Even the notoriously ugly Aztek was able to sell "okay" at a discount to similarly equipped cars.

I guess nobody wants to be the discount brand in the long term...


Great as long as you expect only 50K miles out of them.

Toyota has a completely different strategy,

My mom had a K wagon and loved it; she sold it after our first July in Virginia though (it had no A/C and sitting in traffic with no AC when the dew-point is in the upper 70s is no fun). It definitely had more than 50k miles on it when she sold it.

Yeah, yeah, yeah, Hilux good because Top Gear, we get it,

Car reliability is just applied sociology. Vehicles last however long people keep fixing them and people keep fixing them as long as they expect them to last.

The 80s Japanese cars that the internet loves to circle jerk about where not any better than their domestic counterparts when put in equivalent hands. It wasn't until the 90s when the domestics somewhat ignored their car lines in favor of the new SUV hotness that the Japanese cars pulled ahead in terms of quality. You have to remember that the Japanese had various levels of partnerships with various American manufacturers at this time. It's not like either group was doing things the other was unaware of.

The K-cars and the other domestic stuff was CHEAP. That means that a lot more of them found themselves in the hands of bottom of the barrel customers. The kind of people who today roll 10k of negative equity into a Nissan Altima financed for 84mo with $0 down. Those are not the kind of vehicles that get garaged, get maintained on schedule and get cared for well. The Japanese stuff got put in slightly more caring hands from the get go (because the people who don't care about spending a few extra grand up front tend to also be willing to pony up for more preventive maintenance) and then 20, 30, 40yr out proportionally slightly more of it is still around. Look at what's on the used market for Crown Vics vs what's on the used market for Grand Marquis if you want to see a great example of this. Even if you exclude cop and taxi vehicles Vics will all be clapped out compared to the GMs because they weren't mostly owned by gray hairs who treated them nicely.

You also have to consider whether a vehicle's customer demographic trades in frequently, trades in infrequently or sells private party. The first couple generations of minivans mostly got bought by rich suburbanites who traded them in after say 5ish yr. Exporters bought tons of them. You'd think they're unreliable because you never see them in the first world but there are tens/hundreds of thousands of happy 1995 Ford Aerostar and 2002 Honda CRV owners in places like South America. The lower class variants of those same types of vehicles say the 1999 Town and Country and the 2004 Santa Fe are still abundant om American roads because the buyers were generally less well off and more likely to sell private party.

Go browse bubble Taurus (3rd gen I think) and Subaru Outback sedans on Craigslist, see what kind of condition they're all in then do the same for their more utilitarian station wagon counterparts (and adjust by relative sales numbers) and you'll see how much the expected use case and lifetime of a vehicle is a self fulfilling prophecy. Sedans get used hard as commuter cars and then junked or sold depending on age. Wagons are useful so wagons stick around as long as they run and keep getting fixed as long as it's not something crazy.

One should also keep in mind that all the production stuff from the article only really applies to the first slope in the bath tub curve. Once the vehicles have been on the road 5yr they will either be fixed because they weren't built right or they will have been built right from the get go. How long the vehicles last after that is mostly a reflection of the design rather than the execution on that.

Source: Former low end car flipper, junkyard employee, and currently own a small fleet of random 80s-90s foreign and domestic non-enthusiast cars.

> Vehicles last however long people keep fixing them and people keep fixing them as long as they expect them to last.

That doesn't explain why American cars have way more faults already in the first five years compared to Japanese cars though.

While I generally agree with most of your sociology observations, I feel you're heavily discounting the real impact to reliability that relates to design and specification differences between the domestics and Japanese auto-makers. In the 90's, I was a process engineer for an automotive wiring manufacturer supplying to both domestics and Japanese harness makers. The Japanese clients mandated cold-weld free segments for use in all of their cars with no exception. The domestics were ok with <=2. A cold-weld (physically crimped) going through hot/cold thermal cycles will eventually break in a car and is a messy quality risk because wiring controls a huge amount of car function and depending on where the weld ends up and whether it is a partial or complete break - can result in hard to diagnose failure symptoms (e.g. flaky lights, complete outage, etc ...). We used to charge more (which the Japanese clients happily absorbed) for the weld-free wire because the process lines would need to either be re-strung up for each pass (lower line utilization/efficiency) or we had to physically cut out the weld in the wire based on the detector's marking. This small difference in specifications and acceptance criteria will result in reliability differences downstream. I have no doubt that other suppliers could tell you equally divergent requirements which all may feel small in isolation, but amount to large reliability concerns for the system overall. While I understood the cost-control desire of the domestics, it always felt like a short-sighted decision to accept this type of risk systemically. It no doubt helps to have caring owners, but the reputation that draws them is based on design and specifications made early in the product's design life cycle.

Interesting Side-Note: When I first started, there was an anecdote/story about a weld that had made its way accidentally (weld detector inadvertently turned off for the wrong run) into a production lot. This lot was used in a late design life cycle batch run (~50K cars) for one of the large Japanese manufacturers who were planning on announcing a new design model year for the upcoming year. When the auto maker heard that this weld had made its way into the batch run, they gave our company two choices: Either buy the 50K cars or pay for the auto manufacturers engineers to comb through each car to validate the wiring was weld free. I don't know which choice was made, but the there were alot of human and mechanical systems put in place to make sure the weld detector and the client requirements were well marked and understood for all runs thereafter. It was a fireable offense to mess with that detector in any unauthorized way.

That's a hilarious story about the response to the potential of a defect.

Good observation,

But I want to add that Toyota can mean a dozen real manufacturers under a single brand.

Even in Japanese domestic market, there are "Junk" Toyotas, and Good Toyotas. There are good domestic market Toyotas made by external contractors, and there are bad Toyotas made by Toyota's own factories.

There are export grade Toyotas, made with export grade parts, and there are domestic market Toyotas with significantly different in design, and parts despite being sold under same model name.

And outside of Japan, there are many, many dozens of contract assembly plants ran by completely random people, assembling cars from very wide variety of parts, and doing many design, and spec changes.

By the way, what's the tool is there to detect cold welds in wiring?

An educator who manages a community college auto training program told me the most reliable car is the Toyota Yaris. It looks cheap, is cheap but because it's built to similar standards and has fewer parts it just doesn't break much.

the reliability of a car relates to how long something lasts while not doing fixing (normal maintenance can be expected). so your premise is wrong right out the gate.

This is a very American point of view but you forgot that Japanese and other cars are sold all over the world and usually sold as cheap middle class type of car. Still, their reputation is the same all over the world where people don't maintain them more than they maintain a ford and certainly not maintained as premium European brands bought by upper middle class like BMW or Mercedes.

You touched the subject of design and this is also part of the reason, Toyota sells the largest number of cars in the world and have huge amount of data which they use to perfect the design and fix weak points, this helps a lot to make their cars last longer than other vehicles in the same price range and category.

That may have something to do with the consumers (who are influenced by marketing). People want to own expensive cars because that way they can feel they are better than their neighbor. Car -ride is a thrill especially in a fancy car. No?

I suspect that such extremely optimized design won't be accepted by consumer unless many enthusiasts exists like Tesla.

This is so clearly the optimal organization for a productive firm. But it falls on deaf ears every time. I gave up trying to proselytize. We're going to plan the whole product at the beginning of each quarter, save technical debt for later, over promise and under deliver. But why oh why are we burnt out working the weekends?

This is my experience as well at a large aerospace manufacturer. When push comes to shove nobody wants to stop the line or take the extra time to actually do anything management preaches. TPS is cargo-culted like mad.

If you produce physical goods of any kind, Shigeo Shingo should absolutely be required reading. His ideas around quality control and setup times still sound revolutionary even now.

Can anyone comment here on the "Uddevalla concept"? Quoting https://ideas.repec.org/p/pra/mprapa/10785.html :

> Both researchers and practitioners in industrial organization ask themselves today whether lean production is the only possible model for the future. Enriching Production proposes a radically different alternative, which was put into practice at Volvo’s Uddevalla plant during its brief life span.

The book "Enriching Production" is at https://mpra.ub.uni-muenchen.de/10785/1/MPRA_paper_10785.pdf .

As I have no experience with lean nor with automotive production history, I find it hard to interpret.

I worked in automotive manufacturing at a tier 1 supplier.

I think this is easier to read than the “Enriching Production” book: http://research.engr.oregonstate.edu/gwt/biblio/enriching-pr...

I think there are two differences one is a technical / engineering difference and the other is aimed at the worker enjoyment and well-being. The Toyota (lean) approach is presented as being all about manufacturing efficiency at the expense of the worker where as the Uddevalla approach appears to be sacrificing production efficiency for the benefit of worker enjoyment and well-being.

From the technical side it seems to be the assembly line production approach (lean / Toyota) vs craft / one off production approach (Uddevalla).

Current automotive manufacturing practices are based on an assembly line. This dates back to Henry Ford [1] [2]. The frame of the car travels down a line and components are brought to the workers to be added to the frame. The worker has a specialized task (e.g., attaching the passenger side door) that is repeated for the entire shift (or some specified amount of time).

There used to be a “craft” approach to manufacturing [3]. One person (or a few people) would make the entire product from beginning to end. This seems to be the team approach to production at Uddevalla.

Note: I didn’t read Enriching Production word for word and I have never heard of the Uddevalla approach before.

[1] https://en.m.wikipedia.org/wiki/Highland_Park_Ford_Plant

[2] see the mass production section for an earlier assembly line. https://en.m.wikipedia.org/wiki/Venetian_Arsenal

[3] https://en.m.wikipedia.org/wiki/Craft_production

That summary was much easier to understand, thanks!

One of my complaints about "agile" is that it's nearly all-encompassing. If it isn't strict waterfall then it's agile, in one way or another, so "agile" becomes a pretty useless term.

And I don't much like agile because many of the features seem to have the goal of (to re-use your apt language) removing craft from software development.

I get the same feelings about what little I know of "lean". Everyone loves and claims to use it, one way or another. Which med me wonder about alternatives that aren't followed.

So I was receptive when I came across mention of the Uddevalla approach. I would really like to have seen that approach developed and tested further.

You’re welcome!

I don’t think there is such a thing as strict waterfall. Waterfall vs agile (or even spiral, or V shaped development) arguments don’t make much sense to me. IME, everything ends up being “iterative” in practice. Everything is a tool, use the tool that works for the given situation. I think principles are important and when evaluating different approaches it’s great to see what’s common to them.

Even the paper[1] that is credited (or blamed) for “waterfall” advocated iteration. Maybe I was aware of agile before reading this and that caused me to interpret it differently? That’s possible, but there is mention of prototypes, looping back, and even involving the customer!

I think it's better to think of it as the Toyota Production System (TPS) instead of lean. Lean comes from the MIT study of Japanese manufacturing where they specify 5 steps [2]. This is really the just-in-time pillar of TPS [see the "house" in 3] and there is a lot needed to be able to get there. TPS is an entire philosophy to run an organization. Everything from annual planning at the executive and leadership level all the way down to problem solving at the machine operator. The goal of TPS [shown across the roof of the "house" in 3] is “highest quality, lowest cost, and shortest lead time”.

To me, the core of lean is about problem solving. This can follow anything from the Plan-Do-Check-Act cycle[4], OODA loops, six-sigma's DMAIC[4], Toyota's A3 method[5], and many others I am not aware of or am not including. The core of each of these is usually the scientific method. Again, these are all tools. Sometimes these are not needed, sometimes they are overkill, sometimes they are useful. Everything else in “lean” / TPS is either specific tools (e.g., 5-whys) used to identify problems or best practices (e.g., kanban, SMED) aimed at solving problems.

To give you an example of how we focused on problem solving, we introduced a daily production meeting to a department at the manufacturing plant. The production meeting took 5-15 minutes at the beginning of the shift. The focus was on SQDCME (safety, quality, delivery, cost, morale, environment) metrics. The leads and production planners check-in more often to hit their production targets. After instituting the production meetings and starting to track data more closely, we taught the production team the 7 QC tools[4] along with the A3 problem solving process[5]. We had really good results using these basic tools to solve production problems. These production meetings were driven by numbers, as we had to hit daily targets. If we didn't hit production numbers, it becomes why didn't we hit them? If we are producing too much scrap, it becomes why is there too much scrap? The team is cross-trained on different processes and machines. Often we would grab some cardboard from the recycling bin put it up beside the production board and start going through the A3 process. We scraped 5 of part ABC, here are the physical pieces to inspect that we took from the scrap cage. The work order said it was scrapped at machine 123 in process X, what could have caused this? From there, we can quickly note down a bunch of the hypothesis generated by the group, then we can develop experiments to test these, again, these come from the group. After that we can develop ways to correct and prevent these issues from coming up again. This isn't about "management" telling what they thing the problem is, or how to perform the process. This is about finding a better way and continuing to improve the current knowledge. Good ideas can come from anyone.

Without the ability to solve problems you're not able to start building the foundation of the Toyota house. The foundation refers to stability. This is often stability in the 4M's (that's the old literature): material, method, measurement, and person. How can you ensure stability if you can't solve problems, which is why I try to focus on that and build up the rest as needed.

I'm not sure what you mean by craft of software development. I am currently working as a business analyst in a team that develops in-house business automation software. From what the developers have told me, craft is making nice interfaces, writing clean code, writing reliable and fast software, solving business problems, and making the user of the software happy. When requirements are always changing, or edge cases are not covered, that leads to changing architecture, throwing away code, or software that crashes and users say it doesn't work.

To me, structure reduces some of those problems by focusing on the process of creating software. The process of moving from requirements, software architecture / design, development, testing, and deployment should be somewhat structured. My approach is not to design and document the entire software project before any code is written. I like to at least see, to the best knowledge at the time, the high-level of the project. Who will use it, what are the use cases, what systems do we need to interface with, where are the data sources, what are the outputs. This stuff needs to be documented and preferably reviewed with the team before development starts. From here, you're able to prioritize which use case has the most value, you're able to see what is risky, and what if it changes will cause lots of re-work. I can't count how many times I've seen people change their mind as soon as you show them some simple documentation, or say that's not what I meant, or start developing something to find out that they are solving 10% of the entire problem.

Apparently there is a manifesto of software craftsmanship[6]. If that's what you're referring to, I think it aligns with what my coworkers have told me, and I think (hope) the structure I've outlined above helps to allow my coworkers to achieve that.

[1] http://www-scf.usc.edu/~csci201/lectures/Lecture11/royce1970...

[2] https://www.lean.org/WhoWeAre/LEINewsStory.cfm?NewsArticleId... 1. Specify value 2. Map 3. Flow 4. Pull 5. Perfection

[3] see page 5 for the Toyota Production System "House". This document is what I recommend when someone wants to learn about lean. http://artoflean.com/wp-content/uploads/2019/01/Basic_TPS_Ha...

[4] https://www.sv-jme.eu/?ns_articles_pdf=/ns_articles/files/oj...

[5] http://a3thinking.com/index.html

[6] http://manifesto.softwarecraftsmanship.org/

I know of the history of the term "waterfall". I've also heard people say "we use waterfall", meaning (though they might not know it), something more like Rocye's modified waterfall model. But what I rarely get from agile people is a discussion of what is "agile" vs. other approaches. I came of age as a developer in the 1990s, strongly influenced by McConnell's "Rapid Development", which describes a dozen different development styles. RAD of course was one of the influences of the Agile movement, but I never understood what "agile" meant, and the manifesto still doesn't make sense to me, leaving me to think it's tied to issues that come up in large software development projects, while my career has all been in small to mid-size projects with <200 people years of effort.

You've provided some excellent commentary.

With the current pandemic and two preschoolers home from school, I simply don't have the time to follow your links and give them due thought and consideration.

Probably the fundamental blocking point is that I am self-employed, working with at most a small group of people for consulting/contract work. As such, I simply don't have the experience that would help me really understand how these issues affect things.

I will keep this page open, and read those links more thoroughly at some point, but probably not until after massive vaccination and re-opening of libraries, museums, and other places to burn off pre-schooler energy and attention needs.

The really funny thing is after you learn about TPS, and you talk to project managers, product owners, and executives, they all know it too. And none of them will use any of its lessons.

I've found that they basically fear the change and would rather keep doing things the same way even if it results in problems. They think it's safer to try to find a magic bullet that fits into their model, such as going "all-Agile" or "hiring SREs". Anything that isn't a radical change from their hierarchical, waterfall, arbitrary-deadline, afraid-of-deploying-on-friday, drop-everything-we-have-a-great-new-shiny-ball-to-chase bullshit.

Notice the blurred pics being presented as motion blur trying to do seal “secrets”

Continuous/Incremental improvements aka https://en.wikipedia.org/wiki/Kaizen

I've lost my faith in all this Just int Time marketing stuff. I tried to buy a specific Toyota car in Brazil and they said that it would take 6 months to get me one. I don't know which kind of legal bureaucracy created this, but I went with a Honda.

If you want to buy a >>specific<< configuration of any brand, you will have to wait for months - even for Honda...

More important to the consumer, JIT often means waiting weeks for replacement parts to be back-ordered several levels up the chain. It's a mistake to think there are no downsides to this methodology.

Ultimately JIT manufacturing should be seen for what it is: central planning in action. Good for the central planners, certainly, but not for the end users.

> Ultimately JIT manufacturing should be seen for what it is: central planning in action. Good for the central planners, certainly, but not for the end users.

That's kind of funny. JIT is actually meant to be something closer to the market solution many others propose, as opposed to central planning. It's about identifying the local needs and ensuring they're met rather than trying to drive a top-down solution to the production problem.

JIT doesn't mean you should carry no inventory or wait for orders. You should carry enough stock that you sell out of just in time for the replenishment to arrive.

This is one of the conclusions of playing "The Beer Game"[1] which aims to have players simulate a supply chain.

To begin with players try to decrease local inventory to get a better score. i.e. lower costs by holding less. Then when a shock to the system is introduced, and orders spike, the players struggle to meet demand and the system becomes chaotic.

This seems to happen even when players have played the game before. The lesson seems to be that you can't easily do JIT without very good lines of communication or buffer inventory which is expensive.

As I remember it the cute, old metaphor I was taught was that inventory was like water in a river. We aim to lower the water until you just expose the rocks but no more. Then you remove the rocks one by one. If you deplete your inventory too soon you're exposing too many rocks and you'll ground your boat.

1. http://web.mit.edu/jsterman/www/SDG/beergame.html

The practical implementation of JIT production is to get rid of (expensive) warehouses and instead have rolling trucks on streets be the warehouse - you need the trucks anyway to transport stuff, but now you don't have to pay for the space as roads are owned by the public.

This may be the end game of JIT. But Deming and Shingo would have defined quality as "meeting the customers needs" and having them wait for parts does not accomplish that. Don't let the theoretical get in the way of the practical.

Sounds like the drug dealer model, where product is largely delivered to customers as they need it. Large stashes of drugs have additional potential downside, making the model more attractive than normal.

Capitalist race to the bottom in action, with JIT as enabler.

IIRC, one of the classic elements of kanban was also that you could easily setup a buffer at any point in the system, by controlling the number of cards/boxes, with kanban regulating the flow.

I.e. it's trivial to use TPS approaches to setup a "we need X amount of slack" and have its decentralised pull-based system appropriately adjust flow to maintain said slack.

Unfortunately in steady state the extra cost of maintaining slack is seen as unnecessary drag on company performance, and since you can't depend on competition having to maintain similar amount of slack... why be less competetive?

> Unfortunately in steady state the extra cost of maintaining slack is seen as unnecessary drag on company performance

One of the reasons for maintaining slack is that many processes have been repeatedly shown to be on-average more productive when managed with slack in normal operations than without, because the disruption that occurs when processes have no slack and irregular events occur is much more costly than a certain amount of slack. Right-sizing slack makes you more, not less, competitive.

It's quite often not company performance concerns that drive out slack but the short-term performance of the unit at exactly the level the decision is made, and it's typically an eyes-open decision to gamble on unsustainable short-term gains lasting long enough to meet a short-term goal of the decision-maker (often, a bid to get a better job where they are no longer immediately responsible for the function being compromised).

One of the biggest fallacies is there management decisions within a company are consistently, or even mostly, driven by interests of the company as a whole, rather than the individuals making decisions, who often face practical personal incentive structures (especially in terms of career advancement) in which seeking the broad interests of the company and it's shareholders are not optimal.

I'll admit that I might have been a bit too general (and influenced by https://slatestarcodex.com/2014/07/30/meditations-on-moloch/)

Arguably that's a case of internal competition - the short term gamble will pay off showing you as more competitive, and you're not going to be there for the consequences, are you?

That said, I have encountered (more like been in the office when the discussion happened and got rest of the tale) bean counting issues at highest levels of a telco conglomerate where the "international" level was asking why we were spending so much on BTSes, with particular issue being how we stuck sizeable UPSes on them - whose maintenance wasn't cheap.

Ultimately it turned out that other national units (this was level of individual, nation-wide european telcos in one multinational) were either doing creative accounting or just skipping on reliability. But because of that, their BTSes looked much cheaper than ours, and we were getting the flack from multinational level.

Following JIT practices in production does not mean that one has to abolish spare parts storage.

The only ones known to have massive problems with spare part supply issues is Tesla, which is a contributing factor in their high insurance rates (e.g. per https://electrek.co/2017/06/05/tesla-owners-insurance-rates/).

That is not how JIT works (well maybe in some Americanized version but clearly not at Toyota). Specific orders are made with JIT but parts and cars are also pre-produced.

Sounds like a Brazil or dealer problem to me. My new Toyota landed at my dealer on the 31/12-2020. With JIT production (I didn't buy a pre-produced model but picked parts and colours I wanted) and transport by land - sea - land it took way less than that. The Honda you bought was either pre-produced or you tried to buy a Toyota that was a brand new model and you wanted a specific setup.

You got burned because honda is no longer reliable the way they used to be.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact