Wednesday, May 31, 2006

Gemba Keiei by Taiichi Ohno

Topic: Management Improvement, Toyota Production System, Lean Manufacturing

Jon Miller has been posting thoughts on chapters of Gemba Keiei by Taiichi Ohno for quite some time. His latest post in on chapter 23 (of 37): Producing at the Lowest Possible Cost. The series of posts provide great doses of management wisdom. As he said in the first post: "As I re-read this book in the original Japanese, I will summarize the nuggets of wisdom from each chapter in Mr. Ohno's book."

Some great examples, If You Are Wrong, Admit It:

If you fancy yourself a Lean thinker go ahead and test your beliefs the next time you think you are right. If you are wrong, admit it. Make Mr. Ohno proud.

Great idea. The PDSA is a great tool to help test your beliefs (when related to an idea for improvement). Make sure you predict and then test your prediction.

Wasted Motion is Not Work:

Toyota people think motion is work and Ohno struggled to convince people that this was wrong. Human motion is only work when you add your wits to what you are doing, says Ohno.

Toyota Made the Kanban System Possible:

The suppliers were included in the kanban system at the very end "to minimize confusion to our suppliers". Ohno gives an example of another Toyota manager who caused problems for a supplier who was forced onto a kanban system even before the Motomachi factory at Toyota had succeeded with kanban. Ohno makes a strong statement that you should not implement a kanban system with your suppliers before you have successfully implemented the kanban system in-house. This is advice that is seldom heeded today.


Tuesday, May 30, 2006

Lean Software Development

We have posted on the topic of Lean Software Development previously:

Lean Software Development: A Field Guide - the first 3 chapters of this new book are available online. Excellent, recommended for anyone interested in lean thinking ideas.

Lean Software Development: An Agile Toolkit for Software Development Managers by Mary Poppendieck and Tom Poppendieck, 2003.

Articles on lean programming by the Poppendieck's

David Carlton has an interesting post on this topic: lean software development.

I'm not sure what concrete effects it will have on my work in the short-term; for now, I'm going to continue reading about lean, waiting for concepts to sink in and crystallize. I imagine that I'll return to this book in a couple of years and find specific inspirations in it, and indeed that it will have subconsciously inspired me in the interim.

Software development can be extremely complicated and can benefit greatly from making problems visible - jidoka. Inventory is not the key to hiding problems in software development. It might be in the "production of software" but so much software is "produced" and distributed over the internet without producing inventory etc. that the production step is not the main source of problems and longer (in the past boxed software had the same inventory problems other industries experience today). Yet all of us using software have many experiences with the problems users have with software (often on a daily basis). Lean thinking has a great deal to offer for those involved in software development.

By the way, if you are interested in lean application development and are looking for a job in the Washington DC area see, job announcement for a programmer, and maybe we can work together - John Hunter.


Thursday, May 25, 2006

Improving Engineering Education

Topic: Management Improvement, Engineering Education

On our Science and Engineering blog I just posted on the Olin Engineering Education Experiment. It is a great story of doing things differently.

The Franklin W. Olin College of Engineering was founded with a donation of over $400 million and opened to students in 2002. All students get a full tuition scholarship. Interesting article: The Olin Experiment by Erico Guizzo gives an excellent overview of the different focus of the school:

Olin's aim is to flip over the traditional "theory first, practice later" model and make students plunge into hands-on engineering projects starting on day one. Instead of theory-heavy lectures, segregated disciplines, and individual efforts, Olin champions design exercises, interdisciplinary studies, and teamwork.

To some extent this is something a number of schools are attempting to do. One, of many examples - Princeton University: "At the same time, the center is improving students' technical education by exposing them to real engineering projects throughout their four years, through internships, entrepreneurial opportunities and multidisciplinary courses." - Princeton Center for Innovation in Engineering Education). The nature of Olin's methods do seem to be a qualitative different, not just a matter of degree.

In most traditional schools, students sit through separate calculus, physics, and chemistry lectures during the first two years and have only a few canned-type laboratories. Olin doesn't eliminate each and every "chalk and talk" lecture; some professors do teach that way. But Olin's curriculum, unlike conventional ones, tightly integrates the basic disciplines with practical projects.

This requires radically changing the normal university education model. To me this is definitely a different versus better (see last post) improvement effort. It will be interesting to see the success they achieve going forward. It almost makes me want to go back to school.

Building a Better Engineer by David Wessel:
"If they become another good engineering school, they will have failed," says Woodie Flowers, an MIT professor advising Olin. "The issue is to do it differently enough and to do it in way that will be exportable" to other colleges.

That goal increases dramatically the potential benefits. If they truly focus on not just doing a great job at their school but exporting good ideas to others the benefits of success will be multiplied.

Wednesday, May 24, 2006

Better and Different

Toyota: Better or Different?,Lean Blog commenting on Seth Godin's post

The answer, as I see it, is to be better and different (when necessary). In Seth's post he talks about challenging people to find not just better solutions but different solutions. That is fine, as long as people don't lose focus on being better. Neither one alone is adequate (at least not always). To achieve great success you must be both better and different. That is what Toyota does.

Frankly, if you have to choose one, just being better will work most of the time. The problem is (using an example from Deming, page 9 New Economics) when, for example, carburetors are eliminated by innovation (fuel injectors) no matter how well you make them you are out of business.

Often people mistake Deming's ideas as only about being better. He stressed not only continual improvement (Kaizen, incremental improvement, SPC) but also innovation. He stressed innovation both in the normal sense of innovating new products for customers and also innovation in managing the organization.

If innovation will be poorly executed (because your organization doesn't do things better - just differently) you can buy some time until others can adapt to the innovation, but that is all. Unless you also do things better you will not succeed for long.

I see so many examples of failing to practice obvious better methods: methods that have existed for decades. While, at times, the best outcome would result from finding a different way don't fail to make sensible improvements in the meantime. At least adopt the better methods that are known (doing so is fairly easy compared to inventing new ways of doing things) until different ways are adopted.

Toyota is a great example of doing both. So are Google and Apple. But doing things differently also means taking risks and Apple has suffered in the past. Doing things differently is great as long it is the right differently (which isn't always easy to judge).

Tuesday, May 23, 2006

Financial Education

Topic: Investing

"Financial education is a critical component of a robust and effective financial marketplace but it is not a panacea. Clear disclosures, wise regulation and vigorous enforcement are also essential to ensuring that financial service providers do not engage in unfair or deceptive practices," Bernanke said.

Outlining the various initiatives that the Fed already sponsors to boost public understanding of financial matters, Bernanke pledged to keep up the work.

The financial decisions we make have huge impacts on the quality of lives. This blog focus largely on management improvement: in such posts we often mention the importance of long term thinking and systems thinking. When planning our personal financial paths long term thinking and systems thinking (to optimize our long term financial well being given the options available in our individual situation) are necessary.

One advantage over attempting management improvement is when working on your financial plans you don't have to convince others to change: you have the ability to implement your personal financial plan.

Financial information:

Monday, May 22, 2006

Lean forward

Lean forward by Martin Ashcroft (on the recent lean conference by The Manufacturer magazine):

There is also a growing awareness that lean principles should not be confined to manufacturing operations, with almost nine out of ten recognizing their value throughout the entire organization. Action speaks louder than words, however, and manufacturers betrayed themselves somewhat in their answers to a later question about business initiatives, revealing that in practice, less than half had done anything about extending lean principles into business processes.

This seems exactly right. People agree lean should expand beyond the factor floor but actually doing so lags that belief. Still there is plenty of work to do both in getting companies to do apply the most obvious lean ideas as well as extending those companies that are already successfully applying lean concepts.

The most substantial success requires building upon the initial efforts. Truly becoming a lean organization is not something that can be done easily or quickly. And it can't happen unless management stays focused and continues to learn about lean concepts.


Wednesday, May 17, 2006

Problems Caused by Performance Appraisal

Topics: ,

I ran across a great article on the problems created by our common use of performance appraisal today: Unjust Deserts (pdf format) by Mary Poppendieck:

As Sue's team instinctively realized, ranking people for merit raises pits individual employees against each other and strongly discourages collaboration, a cornerstone of Agile practices.
There is no greater de-motivator than a reward system that is perceived to be unfair.

The article does a good job of explaining these, and several more, problems caused by performance appraisal. It also provides some good thoughts on how to manage effectively, including:

While monetary rewards can be a powerful driver of behavior, the motivation they provide is not sustainable. Once people have an adequate income, motivation comes from things such as achievement, growth, control over one's work, recognition, advancement, and a friendly working environment. No matter how good your evaluation and reward system may be, don't expect it to do much to drive stellar performance over the long term.

People are increasingly challenging the notion that we just have to live with performance appraisal systems. As usually, I will make my suggestion that chapter 9 of the Leader's Handbook offers great material on performing without appraisal (and the rest of the book is great too). More related resources:

Tuesday, May 16, 2006

The Customer Knows Best?

Topic: Management Improvement

The Customer Knows Best? Better Think Again by Anthony W. Ulwick

It's important to listen to customers - but not follow their words without skepticism. Ask them to design your next product and you're likely to miss the mark, suggests this Harvard Business Review excerpt.

Excellent point. Some management ideas are pretty easy and straight forward. But many management practices require knowledge and judgment to apply them successfully. Easy solutions may be desired, but, often you must choose between easy and effective (hint, I suggest effective is the better target).

Listening to customers is important but it is not sufficient. W. Edwards Deming made this point emphatically on page 7 of the New Economics:

Does the customer invent new product of service? The customer generates nothing. No customer asked for electric lights... No customer asked for photography... No customer asked for an automobile... No customer asked for an integrated circuit.

The article by Anthony Ulwick explores how to structure the communication with customers in order to develop products that will be successful. I agree with some of the points of the article. I think, like many such article, it leaps to conclusions that are not supported by the evidence presented.

The results of using the outcome-based methodology speak for themselves. Between 1994 and 1995, Cordis introduced 12 new angioplasty catheters and saw its market share in interventional cardiology grow from less than 1% to nearly 10% in the United States; market share approached 18% in Japan, 20% in Europe, and 30% in Canada. Net sales shot up 30%, and the company's $50 million cash position allowed it to reach into new markets.

No, results do not speak for themselves. In most cases there are many factors that impact the results. I see no reason to believe that this methodology was responsible for those results. It may provide an example of the paring of the use of this methodology with success. That may support an argument for the value of this method but that is as far as you can go without showing direct evidence of causation - which would require theory that could be tested.

Enron's CFO was the CFO of the year according to CFO magazine - as were WorldCom and Tyco's. People could assume the numbers at Enron proved what Enron was doing was correct. But it did not prove that. Until we start to evaluate data more accurately we will continue to mistakenly see proof where it does not exist.

Seth Godin recently posted an excellent note on customer focus: But the focus group loved it.

A properly run focus group is great. The purpose? To help you focus.

Not to find out if an idea is any good. Not to get the data you need to sell your boss on an idea.


Focus groups are very bad at that.

Customer focus is important but at times management must ignore the claims of customers. One of the jobs of management is to know which is needed at a given point.

Saturday, May 13, 2006

Invest for the Long Term


Invest Like a Simpleton by Tim Beyers,

Ten years ago, Tom Gardner boldly picked 10 stocks to buy and hold for the next decade.
Even with a tough week in which Silicon Graphics filed for bankruptcy and Dell admitted its business is not at all like it used to be, the Simpleton Portfolio would have returned more than six times your money had you invested on day one. For context, consider this: Over the same period, the S&P 500 gained approximately 135%.

Quite a nice record. is an excellent web site worth reading for investing education (they do force you to provide an email address, which I think is a bad practice for web site usability, but the content on is worth putting up with the bother - just use the email address you have to deal with these types of hassles).

You can track the performance of a "virtual fund" based on my 10 stocks for 10 years (original post - apr 2005). I created the Sleepwell fund on marketocracy (this fund includes occasional adjustments - 4% turnover in the last year): Sleepwell investment results.

Investment links:

Friday, May 12, 2006

Shigeo Shingo's Influence on TPS

Topic: management improvement, ,

A very interesting article by Art Smalley based on an interview with Mr. Isao Kato: Shigeo Shingo's Influence on TPS. For those interested in the history of the Toyota Production System this article provides some excellent information.

Some background on Isao Kato:

As much as anyone alive Mr. Kato knows the history of TPS development from an insider's point of view from the 1950's forward. He also had working relationships with Mr. Ohno even more so with Mr. Shigeo Shingo (or Dr. Shingo as he is known in the west). One of the interview topics I discussed with Mr. Kato related to the historical role of Shigeo Shingo in the formulation of TPS inside Toyota Motor Company. Much to my surprise the role of Mr. Shingo and actual development of TPS according to Mr. Kato has been somewhat mistaken over the years especially in the U.S.

Mr. Isao Kato: By far the biggest area was helping us develop a course
that replaced the Job Methods (JM) part of TWI. Together we summarized Mr. Shingo's material into a training course that we called the “P-Course which stood for production and how to analyze a production process. As I mentioned he trained a couple thousand young engineers and managers over a twenty year period. His influence on these people and their subsequent ability to see problems and waste was quite large.

The article does convincingly argue those most responsible include the Toyoda's and then many others inside Toyota such as Ohno.

First the origin of the term JIT was coined by Kiichiro Toyoda in the 1930's and not by either Mr. Ohno or Mr. Shingo. Mr. Ohno experimented on machining lines in Toyota and arrived at a working model of TPS with replenishment pull and other techniques before Mr. Shingo arrived on the scene at Toyota.

Additionally the second pillar of TPS of Jidoka dates back to 1902 and the invention by Sakichi Toyoda of his automatic loom that stopped at the sign of a defect. Mr. Ohno came from the automatic loom factory part of Toyota and started separating man from machine and building in quality as soon as he transferred to Toyota's automotive company in 1945...

Thirdly the concept of "respect for workers" inherent in the system is a concept that comes more from the Toyoda family than any other source.

To me (though this may be more my view than the view of the article) the details provided make clear that many people participated in creating the Toyota Production System as we know it today. The article reinforces the role of the Toyoda's (which the recent focus on Ohno and Shingo has downplayed).

However the Toyota Production System as we know it today would exist with just the effort of any several people: not the workers (even Toyota's), not the senior leadership (the Toyoda's), not internal management experts (say Ohno), not engineers (trained by Ohno, Shingo and others), not consultants (say Deming and Shingo), not previous manufacturing experts (Henry Ford). Each group, and the remarkable people involved, participated in making TPS what it is today.

Art: How famous is Mr. Shingo in Japan?

Mr. Kato: Unfortunately not very much. I think it is analogous to the situation with Dr. Demming for example. In the U.S. for many years Demming was ignored and yet widely received in Japan. We invented a famous prize for him. In Shingo's case he is not well known in Japan especially compared to Mr. Ohno. But I believe that Mr. Shingo is somewhat famous in the U.S. and I heard there is even a prize his name.

Art Smalley includes a great deal of additional interesting information.


Thursday, May 11, 2006

Management: Geeks and Deming

Why Business Needs More Geeks by Robert May:

then along came Wall Street. Obsessed with quarterly profit increases and seeing them as disconnected from value creation, Wall Street encouraged businesses to think short-term. The things that led to value creation - things like innovation, continued learning, employee development, long-term focus - were replaced by pump-and-dump management styles.

There is a great deal of similarity between this article and Deming's ideas. Several of Deming's 14 obligation of management and 7 deadly diseases are noted in this quote, including: "Remove barriers that rob people of joy in their work" and the disease - "Emphasis on short term profits." Deming was a physicist so that may explain the similarity of this ideas to geek management culture.

Points from the article:

      "Geeks seek knowledge for it's own sake" - Deming's point 13 "Institute a vigorous program of education and self-improvement." Deming encouraged organizations supporting education of employees - even when unrelated to work in any direct way.

      "Geeks like to experiment" - many of Deming's ideas focus on this point, most obviously is the emphasis on PDSA

      "Geeks openly debate the merits of technical ideas" - again many of Deming's ideas focus on the importance of focusing on actual merits versus people's assumptions or organizational power struggles (though this point might be less direct than some of the others). Understanding variation (also see tampering) is an effort to get people to focus on merits versus arguing over misperceptions or less important details.

      "Geeks are concerned with doing good work just because" - Deming: "Remove barriers that rob people of joy in their work." Geeks have a strong tendency to hold onto this desire even in the face of Dilbert bosses - where many others give up (and even convince themselves they have no such desire - in my opinion as a coping mechanism). Deming (and I) believe everyone has this desire - though I believe many non-geeks have given up hope of having pride in their work.

      "Geeks are about results, not office politics" - I don't see any direct tie here, but the sentiment would support many of Deming's points about how management has lost its way, failing to focus on the important business needs.

More Deming on Management.
Previous Posts mentioning Deming.

Tuesday, May 09, 2006

Understanding Data

Topic: Management Improvement

Statistics Abuse and Me by Jay Mathews:

the Simpson's Paradox numbers. The national average for the SAT went up only 4 points between 1981 and 2005, but the average for whites went up 10 points, for blacks 21 points, for Asians 37 points, for Mexicans 15 points, for Puerto Ricans 23 points and for American Indians 18 points.

How can that be? Is it important? First, yes it is important. Effective use of data is an important part of management improvement. Emphasis the effective, not the data. Use of data by itself is not sufficient.

To be effective you need to learn to think about not what is printed on the page but what lies behind the numbers you see. The numbers are just proxies for the real situation. Look beyond the numbers you see to what they mean and understand how the numbers presented may not fully capture the important details you need to consider.

Ok back to how the SAT numbers could seem to go up fairly significantly for all the sub groups of the total but only a little bit for the total. This numerical quirk is known as Simpson's Paradox. If the proportions of the subgroups (Asians, American Indians...) change then the overall average is effected not just by the changes of the subgroup average SAT scores but the changes of the weighting of each subgroup (so if the overall average is less than even the lowest gain for a subgroup you know that the subgroup weighting must have increased for one or more of the subgroups with lower SAT scores).

Take care when you are make decision based on your understanding of data to avoid assumptions that may not be correct.

Monday, May 08, 2006

Respect for People

Topic: Management Improvement

A very thoughtful post, Respect for People on the Kaikaku blog raises some interesting questions. What does respect for people really mean?

Toyota empowers people: To stop the line - to stop every other worker from working - that is real respect and trust. To implement creative improvement ideas around their work area. They trust you to come up with the best idea to make your work easier and more interesting. You don't have to wait for management to tell you what to do. By asking people to solve problems and become problem solvers.

Those are indications of respect. The post also notes "Ohno was absolutely ruthless, employees and suppliers lived in fear of him." I would say that while Taiichi Ohno was truly remarkable that doesn't mean he did everything right. And he might well have failed to communicate in a way that conveyed respect for people. That doesn't mean doing so is good. It might mean that if you offer as much positive value as Ohno did people may be more forgiving of your weaknesses (I know that is my tendency).

The difference between respect and disrespect is not avoiding avoiding criticism. In fact often if you respect someone you can be much more direct and critical than you can with someone you treat as though they don't have the ability to listen to hard truths and improve. I think we often have so little respect for people we just avoid dealing with anything touchy because we don't want to risk they won't be able to react to the issues raised and will instead just react as if they have been personally attacked. It may also be that it is easier to train managers to behave in this way than in effectively dealing with though issues. But that is not training them to respect people it is your organization accepting you don't respect your people (managers and others) so just train people how to behave in a way that avoids difficult areas.

One way to communicate respect is when you are critical to criticize the behavior or action not the person. When you criticize a person's self instead of an action it is very likely to be taken poorly.

I don't feel I have the ability to present how I see in a way that is easy to follow. But I will try. It is possible to separate respect (which is the state of mind on the person [say management]), communicating respect (which is the ability, of say the manger, to communicate that respect [their state of mind] to others) and communication of content (say the message the manger wants to deliver).

Without respect one way to communicate content without making people feel they are not respected is to avoid any difficult areas and avoid being critical. Avoid confrontation. Speak in the passive voice and act as though difficult decisions are not choices but things imposed on us. It can see that there is respect because no disrespect is verbalized but in truth avoiding criticism is not the same thing as respect.

It is possible to communicate content in a way that is very critical and demanding and yet maintain respect for people. In America this is becoming more and more difficult because, it seems to me, we confuse respect with avoiding anything that might possibly be taken poorly by someone else. This complicates how you communicate respect (though not the actual feeling of respect itself).

Managers in the West normally tell people what to do and rarely ever ask them and listen to their ideas. Listening and empowering people to implement their own ideas are the key to real respect. And develops and educates them - they continually train you on the job and will pay for your college education.

Like so many seemingly simple ideas, "respect for people," is not nearly as simple as it may seem.

Saturday, May 06, 2006

Energy Future

Topic: economics, investing

Interesting chart from: The Oil Age Poster. There are all sorts of opinions on the future price of oil.

My view is based in the capitalist/market model - I believe that if it becomes obvious we are running out of oil the price will go up drastically. Those who own it will feel if you don't buy it today they will sell it to you for much more later. And those that want it don't have much choice (as the last several spikes in oil prices show - demand does not decline without enormous price increases).

Huge price increases will provide incentives to those in the market to innovate to find alternative ways to make money by providing usable energy sources. If the market, overall, chooses to look forward over a long period of time, then investment in alternatives will begin in earnest early and prices of oil will slowly rise. And as prices rise slowly new alternatives (including ways of reducing consumption) will slowly come into the market. Those alternatives will slowly substitute for oil as a smooth transition is made.

If markets actual were efficient and driven by looking far into the future and discounting cash flows to the present this slow and steady model is what would happen. The market can only be efficient if good long term predictions can be reflected in the market. If for example, there is proof we will run out of oil soon but those with the ability to effect the market did not understand this data the market can remain inefficient (the market can fail to reflect the information that is available in an efficient manner leaving room for those with the knowledge to gain above market returns on their investments).

On the other hand if there is no real knowledge (say with a chaotic system where it is impossible to predict far into the future) someone can still "predict" and they could be right by pure luck. A market failure to include that would not be inefficient there was no true prediction that an efficient market should have incorporated. In efficient market theory those with knowledge exert a sufficient impact on the market to make the market efficient.

Markets often take wild and crazy swings with little change in long term prospects which is only reason to question the validity of pure market efficiency. Also, short term pressures lead people to take action for short term benefits (say selling their oil today even if they believe they could sell it for much more 5 years later) instead of making rational long term economic decisions. In a case with a less than purely efficient and less than perfect knowledge of the future, the market will still work to provide alternative energy sources but it will be much more chaotic.

In this case, instead of many investors achieving reasonable ("market") returns as the market behaves rationally they will get huge returns if they are right and huge losses if they are wrong. Those that are right in investing in energy alternatives will make huge returns as the market as a whole waits to invest until the evidence is beyond question (instead of anticipating based on long term thinking and long term investing) and chaotic spikes occur. If that happens the early investors will find a ready market for their alternatives in the market and make a huge return before competitors can react. Or those early investors will lose huge percentages as they invest in alternatives and they lose as oil stays cheap.

Good luck figuring out what will happen. If you are right and invest with some intelligence you can make quite a bit of money (as you can see below, I invested some with the expectation that energy prices would increase).

Related Posts:

Friday, May 05, 2006

Lean Thinking Misconception

Topic: Management Improvement

Office Furniture Companies Now Leaner

Steelcase, the largest of the nation's office furniture companies, cut thousands of jobs, consolidated manufacturing operations and started relying more on third-party suppliers and outsourcing as it shifted toward a leaner and more flexible manufacturing model popularized by Japanese automaker Toyota Motor Corp.

This does not convey true lean thinking. It is a shame so many think lean is equal to "reducing staff." Lean is about removing waste. Removing waste might mean that fewer people are needed but no real lean thinking organization aims to reduce staff. It may be that, due to past practices and current realities, they have no reasonable alternative but to reduce staff.

But, it would be much better to redeploy staff not needed for streamlined processes to other jobs to service the companies growing business. The popular press continues to write as those the main goal of lean is to reduce staff.

Lean is named for the Toyota Production System. You don't see Toyota laying people off they are growing and hiring people. That is what successful lean companies do.

Wednesday, May 03, 2006

Find the Root Cause Instead of the Person to Blame

Topic: Management Improvement

When encountering a problem or defect the inclination of many is to find a person to blame. W. Edwards Deming believed that the system was responsible for 93% of the problems and over time he increased that number to at least 97%. Why did he see it that way, while so many others first inclination is to blame someone?

As I see it the issue has to do with what is the effective way to improve. Often if you ask why do we have this problem or defect, people will point to some error by someone. So you can blame that person (there are reasons this is not a very accurate way to view the situation often but even without accepting that premise the blaming a person strategy is not wise). The reason the blaming a person is a bad idea is that your organization will improve much more effectively if you keep asking why.

Why did they make that error? Why did the process let them make that error? When you follow the why chain a couple more steps you can find root causes that will allow you to find a much more effective solution. You can then pilot (PDSA) an improvement strategy that doesn't just amount to "Do a better job Joe" or "that is it Joe we are replacing you with Mary." Neither of those strategies turns out to be very effective.

But investigating a bit more to find a root cause can result in finding solutions that improve the performance of all the workers. What kinds of things? You can apply poka yoke (mistake proofing) concepts. You can institute standard practices so that everyone is using the best methods - not whatever methods they have developed over time. You can rearrange the process to simplify the steps and eliminate chances for errors. These improvement, and many more, are sustainable and can be built upon over time.

In addition, the psychology effects of seeing people as the source of errors and defect instead of seeing people as the source of improvements to process weaknesses are powerful. If you find yourself thinking a problem or defect is the fault of a person try asking why a couple more times and see if you can find a system improvement that would eliminate or mitigate such problems in the future. That is a much more effective improvement strategy.

I always have had a bias toward finding system improvements but over time that bias has increased as I have applied management improvement concepts. As you gain experience working on improving systems you gain experience showing the wisdom of Deming's 93-97% figures. My belief is that he increased the percentage of problems attributable to the system over time as he experienced the same thing.