Sep 21 2020

Does Toyota Use SPC?

As part of a discussion started by Lance Richardson on LinkedIn, I stated as a fact that SPC was not part of the Toyota Production System (TPS), which prompted several contradictors to tell me I didn’t know what I was talking about. The “evidence” they provided, however, does not refute my statement. It confirms it instead.

Contents

What is SPC?
SPC on the Toyota History Website
Quality in a Toyota Forklift Plant
For the Curious

What is SPC?

By SPC, I meant the body of knowledge currently taught by the ASQ and as part of Six Sigma training, and centered on Control Charts, as developed by Walter Shewhart in the 1920s and refined until the 1940s. It includes none of the data science developed since the computer was invented.

SPC on the Toyota History Website

As evidence that SPC is part of TPS, Steven Bonacorsi directed me to a page on the Toyota history website describing the company’s actions regarding quality from 1949 to 1965. That it is the most recent reference to SPC by Toyota does not say anything about 2020.

You can find the most recent picture of a Toyota employee working with a Control Chart on Art Smalley’s website. It is from the 1950s. According to him, by the time he worked at Toyota, in the 1980s, “these charts were essentially gone.”

Quality in a Toyota Forklift Plant

Then Robert Testerman instructed me to watch “SPC applied all over a Toyota forklift plant” in a Youtube video from 2016:

Perhaps he should take his own advice and watch this video. About half of it is about quality but none of it about SPC. It’s all about all 900 employees inspecting their own work and checking every part. There is not a single control chart in sight.

Toyota achieves quality with methods that have nothing to do with SPC, and it is misleading to promote SPC by describing it as part of TPS in 2020. The keywords are the following:

One-piece flow, for rapid problem detection
Rapid response
Successive inspection, with each operator checking the work of his or her predecessor
100% go/no-go checking
Mistake-proofing/Poka-Yoke
…

The approach is now called JKK, for Ji Kotei Kanketsu (自工程完結), which literally means “autonomous process completion.” It works in automotive; it may not in semiconductors or frozen foods.

For the Curious

Several reactions to this post on LinkedIn meant “curious.” This topic has been a concern of mine for some time, and I have written articles and blog posts about it over the years that may satisfy the readers’ curiosity:

And Renaud Anjoran also chimed in:

Where Error Proofing Fits in a Quality Improvement Approach (2017)

#toyota, #tps, #spc, #jkk

By Michel Baudin • Data science • 30 • Tags: JKK, SPC, TPS

30 Comments

Robert Edward Cenek
September 21, 2020 @ 2:25 pm

Interesting….surprised me initially. I do recall visiting the Georgetown plant about years ago and noticed that 3-4 employees visually inspected the outside of each vehicle at the end of assembly.

Reply
Geoff Elliott
September 22, 2020 @ 12:38 am

but Toyota (TPS) includes fits, finishes and tolerance (parts and assemblies) and metrology

Reply
- Michel Baudin
  September 22, 2020 @ 4:12 am
  
  Critical dimensions, tolerances, metrology, etc., are legacies of interchangeable parts technology that everyone uses. SPC is a specific set of tools to analyze data within this context, and JKK is a more powerful alternative for the automotive context.
  
  If, on the other hand, you need to make a high-technology process capable, you need data science but data science in 2020 and the tools SPC don’t cut it. They predate the invention of the computer.
  
  Reply
DJ Duarte
September 22, 2020 @ 12:45 am

Hi Michel, I absolutely agree that SPC is not a prescribed method of engagement within the TMS/TPS system. I’ve travelled to 16 different Toyota sites here in Japan, there are 31 (10 of the major 12) and I’ve never seen “control charts” of any kind on the shop floor.

However, I did once see in the Quality Control section what looked to be using a “run chart” that captured the number of a very specific quality issues occurring on the left trim of the Daihatsu Move.

Most of the production systems rely heavily on visual management, andon utilization when abnormalities are identified (not to standard) and rapid problem solving capabilities (which include escalation mechanisms.

Anyhow, I’m just another TPS guy who believes what he preaches as I’ve seen and done a lot in my 32 years of optimization. Cheers for now Michel.

Reply
Kyle Harshbarger
September 22, 2020 @ 6:21 am

It is interesting that SPC is not used by Toyota because they’ve found better ways to achieve controlled processes. Does Toyota have any use for auditing?

Reply
- Michel Baudin
  September 22, 2020 @ 7:11 pm
  
  Good question. I think so but I don’t know it for a fact. I would rather let a Toyota alum take that one.
  
  This slide set from Toyota Alabama may help:
  https://aama.memberclicks.net/assets/docs/toyotas%20perspective%20on%20quality.pdf
  
  Reply
  - Kyle Harshbarger
    September 24, 2020 @ 6:48 am
    
    That’s a great document! I found a couple auditing activities in there:
    Assembly Point Management (Slide 27)
    Warranty Audit (Slide 58)
    
    The ASM Audit is the primary one. It looks like a common theme is to audit where continuous improvement is likely to be overlooked or infrequent activities.
    
    It is interesting that management seems to be responsible for auditing, not some kind of outside party.
Raphael Garcia
September 22, 2020 @ 10:58 pm

Comment on LinkedIn:

SPC describes “variation” and “excess” control….is n’t it the definition of muri and mura? Just as when we say reducing muda, we do not necessary think about a specific tool, that could be the management of the limiting constraint for example.

The same for agile: reducing the time to test a software, “do not let a program run when you know it can be repaired earlier”,….it sounds familiar too. SPC (6 sigma great grand father) linked to muri and mura, TOC linked to muda, Agile linked to gemba,.. some how a “way” to take part of a culture into a tool,….Humble opinion.

Reply
- Michel Baudin
  September 23, 2020 @ 10:59 pm
  
  Comment on LinkedIn:
  
  Sign up for a class called “SPC” and you are not going to learn in general about eliminating variation. You are going to be indoctrinated into specific and ancient statistical techniques. You are going to learn Shewhart charts, Western Electric rules and, if you are lucky, cusum and XmR charts. You may even learn about scatterplots and histograms.
  
  You will not learn any of the tools and techniques a company like Google requires from data scientists, some of which are relevant to process control.
  
  Reply
- Narayana Rao KVSS
  September 23, 2020 @ 11:00 pm
  
  Comment on LinkedIn:
  
  Muri and Muda were developed as ideas to implement scientific management in Japan. But some persons, want to people to forget sc. mgnt and IE and think of their tools as the only means of salvation.   I read that Yoichi Ueno is an important figure in developing scientific management and motion study in Japan. He brought the thinking of Buddha and Confucius into efficiency thinking. He introduced the words Muda, Mura and Muri.   https://nraoiekc.blogspot.com/2016/06/yoichi-ueno-japanese-leader-in.html
  
  Reply
David Fleming
September 23, 2020 @ 7:21 am

My experience is that Toyota does not use SPC in their processes. They have a long and successful history of understanding product design and equipment capability in research and development.

Each model change brings new, more capable product and equipment to the floor (continuous improvement). The process engineer works to improve the equipment once it is installed. Once the equipment is in production the checks are primarily done by production personnel with Go/No Go and visual checks.

Equipment design includes error proofing (poka yoke). Problems are identified in-process and production engineers work to find root cause through gemba.

Reply
Michel Baudin
September 23, 2020 @ 10:59 pm

Comment on LinkedIn:

Sign up for a class called “SPC” and you are not going to learn in general about eliminating variation. You are going to be indoctrinated into specific and ancient statistical techniques. You are going to learn Shewhart charts, Western Electric rules and, if you are lucky, cusum and XmR charts. You may even learn about scatterplots and histograms.

You will not learn any of the tools and techniques a company like Google requires from data scientists, some of which are relevant to process control.

Reply
Narayana Rao KVSS
September 23, 2020 @ 11:00 pm

Comment on LinkedIn:

Muri and Muda were developed as ideas to implement scientific management in Japan. But some persons, want to people to forget sc. mgnt and IE and think of their tools as the only means of salvation.   I read that Yoichi Ueno is an important figure in developing scientific management and motion study in Japan. He brought the thinking of Buddha and Confucius into efficiency thinking. He introduced the words Muda, Mura and Muri.   https://nraoiekc.blogspot.com/2016/06/yoichi-ueno-japanese-leader-in.html

Reply
Renaud Anjoran
October 10, 2020 @ 1:07 am

I’d add that Toyota manages the production of its forklifts very differently from that of its cars. They see it more as a low-volume industrial product. Very different teams and general approach, from what I saw.

Reply
Manoel Santos
October 22, 2020 @ 10:51 pm

Comment on LinkedIn:

I really don’t know if Toyota uses SPC or not. Althouhg my Sensei always told me to make sure of 3 points about quality: (i) do not receieve/accept a defect; (ii) do not produce defect; (iii) do not pass a defect to your customer / forward. One day I found an insight in one of Shingo’s book that shed a light on such an approach.

It was about Quality Control. It mentioned that SPC is not the proper way to do Quality Control, as you have to allow several defects to pass on to your customer in order to have enought datapoints and only when considered out-of-control you should act. Most of the time that would mean a root cause analysis with data that didn’t happen near the timeframe that happened the defect.

The basis for Toyota Quality is Jidoka, which focus on stoping defects right at the moment it occurs. Plus Poka-yoke solutions in products and processes plus thousands of inspection points and the andon philosophy, I would say that I would not be surpised if you are correct and find it out that Toyota does not use SPC.

But I would be also very curious on the other possibility as well. And would surely like to learn with what purpose and to what waste SPC is used as a contermeasure there.

Reply
- Michel Baudin
  October 22, 2020 @ 10:52 pm
  
  Comment on LinkedIn:
  
  Shingo’s rejection of all statistical methods, together with Mikel Harry’s embrace of them got me thinking at why these two people held such diametrically opposed views, which they both express as context-free absolutes.
  
  In reality, their thinking was framed by different contexts, Shingo’s by mechanical industries and Harry’s by electronics.
  
  Once you clarify this, they can both be right.
  
  Reply
- David Martin
  October 22, 2020 @ 10:53 pm
  
  Comment on LinkedIn:
  
  Excellent comments Michel and correct. And no TOYOTA does not specifically utilize SPC.
  
  Reply
Robert E. Nugent
October 23, 2020 @ 10:51 pm

Comment on LinkedIn:

ISeems to me the wrong question, the question that should be asked is how does Toyota achieve quality levels the equal or better than their competitors without six sigma programs and supposedly not utilising SPC.

The definition of SPC as being specifically related to control charts also seems somewhat limited SPC is far more than the use of control charts.

I also find it amusing that “facts” are based on what others have said on the internet and you tube videos.

I will defer to responses from current Toyota employees or alumni and suggest going to the gemba and actually seeing what Toyota do.

Reply
- Chris Stergiou
  October 23, 2020 @ 10:52 pm
  
  Comment on LinkedIn:
  
  Practical explanation here:
  
  https://www.whatdadcooked.com/article/the-art-of-perfection-kaizen/
  
  Philosophical explanation here:
  
  https://en.wikipedia.org/wiki/Wabi-sabi
  
  Reply
- Robert E. Nugent
  October 23, 2020 @ 10:54 pm
  
  Comment on LinkedIn:
  
  The SAE manual on SPC contains far more than control charting.
  
  For example design of experiments, process capability studies, Cp, Cpk.
  
  Process capability is an integral part of SPC
  
  When I worked for an OEM you were required to provided process capability studies based on your initial production runs which didn’t involve control charts.
  
  If the references you provide only focus on control charts then they are sadly lacking.
  
  So no “all” the literature on SPC isn’t specifically related to control charts at all and if your going to comment about blanket statements best you don’t base your whole comment/ article on them.
  
  Feel free to answer the challenge as to why Toyota don’t need to use them or Six Sigma yet continuously meet or exceed their competitors quality.
  
  Perhaps we should be learning from them.
  
  Reply
- Michel Baudin
  October 23, 2020 @ 10:55 pm
  
  Comment on LinkedIn:
  
  About the origins of this information that you find amusing, the internet and YouTube are just channels. I am citing a Toyota web page about the history of the company, Art Smalley, an ex-Toyota employee turned consultant whom I have met and respect, and a video from a Toyota forklift plant in the UK.
  
  I have known about this for decades, from personal communications I can’t cite. I am citing these publicly available sources so that it is clear I am not asking anyone to take my word. I appreciate the facts being confirmed by multiple ex-Toyota employees.
  
  Reply
- Robert E. Nugent
  October 23, 2020 @ 10:56 pm
  
  Comment on LinkedIn:
  
  A few points. If Toyota do not use SPC and yet still meets or exceeds its competitors quality why aren’t we learning from them, they do it without a single Black Belt or by your claim without SPC.
  
  SPC is not just control charts, unfortunately I am remote from my references but will attempt to dig them out.
  
  Automotive was still using SPC in the 20 years I worked globally. As recently as 5 years ago.
  
  If your statement was Toyota “don’t use control charts” I would agree.
  
  Reply
- Michel Baudin
  October 23, 2020 @ 10:57 pm
  
  Comment on LinkedIn:
  
  You ask “Why aren’t we learning from them?” Good question! I have been trying my best to learn from them and to pass on what I learned.
  
  As a quality improvement strategy, it is obvious to me that you should use statistics/data science to make processes capable, then move on to using one-piece flow for rapid problem detection and response, and, finally, to Poka-Yoke to eliminate human error. The data science gets you from 30% defectives to 3%, one-piece flow to 0.3%, and Poka-Yoke to 1 ppm.
  
  I presented this concept at an SME Lean Conference in 2001 and taught it with Kevin Hop in our Lean Quality course but I have never seen it in the literature of the quality profession.
  
  Reply
- Robert E. Nugent
  October 23, 2020 @ 10:58 pm
  
  Comment on LinkedIn:
  
  Interesting we are certainly n the same page in that regard. I’m also fascinated if they aren’t using SPC in the machining environment although modern machinery can measure and self adjust in some applications.
  
  We are a long way from the days of having to match engine components for example and mossy engines now are what would be considered blue printed 40 years ago.
  
  I’ve seen some horrific examples of SPC Being misused or ignored including scrapping engine blocks where the bore was oversize and the charts weren’t filled in because the gauge showed we were within tolerance (microns of the upper tolerance)
  
  Reply
- Michel Baudin
  October 23, 2020 @ 10:53 pm
  
  Comment on LinkedIn:
  
  It’s actually explained at the end of my blog post and in Todd McCann’s slideshare on JKK, which is what Toyota now calls their approach.
  
  In all the literature on quality, SPC is specifically related to Control Charts.
  
  In Douglas Montgomery’s Introduction to Statistical Quality Control, Part II, about SPC, covers only Control Charts, in 350 pages. DOE is a separate topic and the subject of Part III.
  
  In Juran’s Quality Control Handbook, 5th Edition, Section 45 about SPC covers exclusively Control Charts.
  
  In the ASQ course catalog, SPC Basics is all about Control Charts. Advanced SPC adds the Normal Distribution as the foundation of Control Limits, and DOE to make Control Limits tighter.
  
  I think it is better for communication if we don’t lump everything under the same name.
  
  Reply
Michael Bremer
October 23, 2020 @ 11:01 pm

Comment on LinkedIn:

Emailed a friend your question about SPC and Toyota. He was plant manager of a 100% Toyota owned supplier operation. Here is what he said (true as of 5 years ago).

….in my day – YES (they did use SPC).

They were used on every line – single piece flow or otherwise – but possibly were not recognisable as such by visitors.

As you would realise Toyota has built a lot of competence into the people – so the format of SPC charts used evolved with the competence of the people.

For example – after two years in the Greenfield startup – everything on a line had condensed down to a single chart that combined everything into a form of a monthly run chart.

So visitors did not recognise it as SPC….there was SO LITTLE VARIATION that traditional LEAN charting was completely pointless. (Showing a flat line chart with zero variation just for the sake of visitors – is just disrespecting the competence of my people).

In this scenario the only time you would see more traditional SPC charts was when we were making a significant change to the line. They would measure the line in more intense detail until they were sure everything was back under control.

Reply
- Michel Baudin
  October 23, 2020 @ 11:02 pm
  
  Comment on LinkedIn:
  
  It would be great to have your friend explain what they were doing. From your account, I can’t tell whether they were (1) plotting parameters over time, or (2) taking samples from batches, computing stats from samples drawn from the batches, and checking these stats against control limits set in a process capability study, .
  
  Plotting time series of critical parameters is generally useful and (almost) everyone does it. If they did (1), they responded to what they saw in these plots, and it may have served them well. Unless they did (2), however, it wasn’t what I have been meaning by “SPC” in this discussion, which is what you learn when you take a class on SPC.  I don’t see SPC as a philosophy but as a specific set of tools. You can pay to get trained to use them and receive a belt as certification from various organizations. If a manufacturer doesn’t use this specific body of knowledge, it’s not using SPC.
  
  It doesn’t mean it’s not using data science to develop and stabilize processes. At a sufficiently high level of abstraction, an abacus and a computer are both digital calculation devices. I don’t think tuition for abacus training would be money well spent.  As far as I am concerned, your friend and his company can call what they are doing “SPC” if it suits them. What I find misleading is teaching WW-II vintage statistics as part of “Lean,” which many will assume to mean “TPS.”
  
  TWI is WW-II vintage too but, unlike SPC, it hasn’t aged a day. The 3×5 cards may have aged but the concepts haven’t, because they are about people. SPC is about data, and it has aged because information technology has changed. 
  
  Reply
Michel Baudin
October 23, 2020 @ 11:02 pm

Comment on LinkedIn:

It would be great to have your friend explain what they were doing. From your account, I can’t tell whether they were (1) plotting parameters over time, or (2) taking samples from batches, computing stats from samples drawn from the batches, and checking these stats against control limits set in a process capability study, .

Plotting time series of critical parameters is generally useful and (almost) everyone does it. If they did (1), they responded to what they saw in these plots, and it may have served them well. Unless they did (2), however, it wasn’t what I have been meaning by “SPC” in this discussion, which is what you learn when you take a class on SPC.  I don’t see SPC as a philosophy but as a specific set of tools. You can pay to get trained to use them and receive a belt as certification from various organizations. If a manufacturer doesn’t use this specific body of knowledge, it’s not using SPC.

It doesn’t mean it’s not using data science to develop and stabilize processes. At a sufficiently high level of abstraction, an abacus and a computer are both digital calculation devices. I don’t think tuition for abacus training would be money well spent.  As far as I am concerned, your friend and his company can call what they are doing “SPC” if it suits them. What I find misleading is teaching WW-II vintage statistics as part of “Lean,” which many will assume to mean “TPS.”

TWI is WW-II vintage too but, unlike SPC, it hasn’t aged a day. The 3×5 cards may have aged but the concepts haven’t, because they are about people. SPC is about data, and it has aged because information technology has changed. 

Reply
Iain Low
October 28, 2020 @ 9:25 am

Comment on LinkedIn about experience at Toyota in the UK in 2002-2004:

I never experienced the use of any complex statistical approaches, and that includes SPC. They use basic quality tools but nothing based on the normal distribution. I did question this at the time with the asst. gen. manager.

The approach is to give the production team tolerances far tighter than the actual required tolerances, knowing that results lying outside this would still generally be fit for purpose (but be investigated for root cause).

This is an interesting concept, particularly compared to Rolls-Royce where the whole design tolerance is given to production to operate within. That meant that when production results fell outside those bands a detailed concession had to be approved, which involved substantiation of the fitness for purpose of the component on a case by case basis.

Toyota improvement works on the principle of defining a standard for things we never realized we needed a standard for. This includes things like inspection of new cutting inserts which I assumed would always be precision made and to a very tight tolerance. But…no. Consumable tooling suppliers make the same cock ups as the rest of us.

Reply
Process Control and Gaussians
March 6, 2024 @ 9:14 am

[…] Does Toyota use SPC? (2020) […]

Reply

Does Toyota Use SPC?

What is SPC?

SPC on the Toyota History Website

Quality in a Toyota Forklift Plant

For the Curious

Like this:

Related

30 Comments

Leave a Reply Cancel reply

Follow Blog via Email

Recent Posts

Categories

Does Toyota Use SPC?

What is SPC?

SPC on the Toyota History Website

Quality in a Toyota Forklift Plant

For the Curious

Share this:

Like this:

Related

30 Comments

Leave a Reply Cancel reply

Follow Blog via Email

Recent Posts

Categories

Social links

My tags