My instructor would like me to clarify: is it the PDCA or PDSA cycle? He thinks Deming changed it to PDSA and is wanting to clarify that.

My instructor would like me to clarify: is it the PDCA or PDSA cycle? He thinks Deming changed it to PDSA and is wanting to clarify that.

PDCA originated with Walter Shewhart, and Deming referred to it as the "Shewhart Cycle." Deming later changed the "C" to "S."

He did.

He changed it because Study implies understanding the sources of variation in the process.

The Act portion is somewhat more controversial, if a four word concept can be controversial. Act should probably mean improvement if the PDCA cycle is used for process improvement, and something like make the changes permanent if the PDCA cycle is used more generally. Or something.

He did.

He changed it because Study implies understanding the sources of variation in the process.

The Act portion is somewhat more controversial, if a four word concept can be controversial. Act should probably mean improvement if the PDCA cycle is used for process improvement, and something like make the changes permanent if the PDCA cycle is used more generally. Or something.

I've always thought that "decide" is better than "act," because what comes after checking/studying should be a decision to either act or leave well enough alone.

Shewhart- PDCA
Deming- PDSA
Wynne- PDSD

Shewhart- PDCA
Deming- PDSA
Wynne- PDSD
LOL - good one.:lol:

Shewhart- PDCA
Deming- PDSA
Wynne- PDSD
Yeah! PDSD, I like this, and have experienced that in some cases it is better to decide to do nothing rather than use "process molestation" just for the sake of doing something. Leave it alone and quit adjusting for every variation.

Why did you ask this when you got the answer 2 weeks ago:bonk: :bonk:
"Dr. Deming’s PDCA cycle is applicable to any process" advice/suggestions? (http://elsmar.com/Forums/showthread.php?t=16029)

The Act portion is somewhat more controversial, if a four word concept can be controversial.

The concept itself doesn't seem to be controversial to me. But some "explanations" of this concept that I've seen in the past few weeks definitely are more confusing than explaining. Sometimes some people are r-r-really good at making simple things look complicated... or, quite the contrary, oversimplified.


Act should probably mean improvement if the PDCA cycle is used for process improvement, and something like make the changes permanent if the PDCA cycle is used more generally.


I think that equating the "act" step with improvement is one of the most frequent mis-interpretations (oversimplifications) of this step - exactly because of the reason mentioned by Jim and Ralph:


I've always thought that "decide" is better than "act," because what comes after checking/studying should be a decision to either act or leave well enough alone.



Yeah! PDSD, I like this, and have experienced that in some cases it is better to decide to do nothing rather than use "process molestation" just for the sake of doing something. Leave it alone and quit adjusting for every variation.


Personally, I am not happy with the name "act", too. But I think that just "decide" would be even less adequate because it would imply just making a decision, without acting upon it. I believe that even a decision to reject (reject whatever had just been planned, done and studied) should be accompanied by some activity (like, for instance, documenting the decision and its reasons, some "cleanup" after the cycle, etc.).

At the same time, I must agree that many interpretations of PDCA or PDSA cycle totally miss the point that the "act" step includes a very important accept/reject decision to be made.

BTW, from time to time, I feel a strong urge to unite the "study" and "act" steps into one. I think one reason of such feeling is that these two steps - together! - are exactly what many people/organizations overlook: they do "plan" and "do", but that's it. Wouldn't it be easier to convince someone that they are missing just one step (out of three necessary steps) rather than two (out of four)? ;)

Wouldn't it be easier to convince someone that they are missing just one step (out of three necessary steps) rather than two (out of four)? ;)

Not necessarily. You'd have to have agreement...

I prefer the separation of phases, because it reinforces the importance of analysis/study/what-have-you before the deciding/acting (which may include deciding not to act, as Jim so wisely points out). Too often people leap straight into 'solution mode' without sufficient analysis beforehand. But of course no engineer would ever do this, I know. ;)

If you understand the method described why does it matter if it is a "C" or an "S". I think it is a silly waste of time for people to change one letter of an accronym and act like it is revolutionizing the world. It is just another form of plagerism in my mind.
Yes Deming changed it but did he really??? Same intent. I have heard that Demings change made it a continuous cycle. How? By changing a word? You make it continuous by deciding that continuously spiraling toward the sun (or ground) is what you need to do. I am unconvinced that the continuous cycle is beneficial.
In most situations once the objective is complete I believe the cycle should stop for evaluation against the Paraeto principle so that the most important and highest value projects do not get lost because of an obsesive compulsion in one area.

Mark

If you understand the method described why does it matter if it is a "C" or an "S". I think it is a silly waste of time for people to change one letter of an accronym and act like it is revolutionizing the world. It is just another form of plagerism in my mind.
Yes Deming changed it but did he really??? Same intent.


IMHO, the names and vocabulary do matter. Otherwise, it would be sufficient to call this particular method "1234" instead of "PD(whatever)A", wouldn't it? As far as everybody understands what all the four steps mean, why bother naming them, right? :)

Name changes do not have to revolutionize anything. If a name change makes the underlying concept better (e.g. more understandable, or more memorable, or more difficult to misinterprete, or whatever) - great! What's wrong with evolution - with improving things bit-by-bit? Most of the time it is actually the best we can do.

And, by the way, I never heard anybody saying that Deming had revolutionized anything just by changing 'C' to 'S' in Stewhart's cycle. All I heard is that he improved understanding of the "Check" step and/or made it applicable more generally. IMHO, this hardly qualifies for a revolution, but it is a nice evolutionary step.


I have heard that Demings change made it a continuous cycle. How? By changing a word?


Whomever you heard this from must have been just another misinterpreter of both concepts.


You make it continuous by deciding that continuously spiraling toward the sun (or ground) is what you need to do. I am unconvinced that the continuous cycle is beneficial.
In most situations once the objective is complete I believe the cycle should stop for evaluation against the Paraeto principle so that the most important and highest value projects do not get lost because of an obsesive compulsion in one area.


That's an excellent point: perfection is not achievable, and endless race towards perfection is probably not affordable by anyone. In practice though, I think all the systems are so far from perfection that the cycle can be repeated for as long as the system under improvement exists. That is, practically continually. ;)

Nevertheless, the essense of you point is still valid. :yes: Each step in this cycle should be justifiable from the cost/benefit viewpoint; and the subject of the next cycle should be selected (among multiple candidates) based on cost/benefit analysis, too. That's why the accept/reject decision in "Act" should include a question like "Even if there was an improvement, was it worth its cost?". And that's why the "Plan" step should include the similar question about the planned experiment (before the costs of "Do" and "Study" are incurred).

But then again, by adding all the details like this to the model - aren't we risking to lose the essense of the model (which I believe is a certain sequence of certain four activities) behind the details? IMHO, it is way too easy to misinterprete or mis-apply as it is...

I prefer the separation of phases, because it reinforces the importance of analysis/study/what-have-you before the deciding/acting (which may include deciding not to act, as Jim so wisely points out). Too often people leap straight into 'solution mode' without sufficient analysis beforehand.


You meant to say "...leap straight into 'acting mode'...", didn't you?

As far as I understand, there is no place for devising any solutions on the "Act" step - that's what the "Plan" step is for, isn't it? The essense of "Act" is to accept (institutionalize, standardize, etc.) or reject the solution that was just planned, tried and studied.


Too often people leap straight into 'solution mode' without sufficient analysis beforehand. But of course no engineer would ever do this, I know. ;)


Well, depends on what kind of engineer you are talking about. :)


:topic:

Now that my original professional background (which, by the way, officially is called "software engineering") was divulged to you by my "boss", I have to admit that I hate the title "software engineer" because in the real world it has very little in common with what most people mean by "engineering". In particular, most of the people I've met who carry this title are wa-a-ay too prone to jump to solutions/decisions without preliminary analysis and study. At least that's what happens when you leave them to themselves. Very little "P" and a lot of "D". And then a lot of "R" ("redo"). :mad:

You meant to say "...leap straight into 'acting mode'...", didn't you?
No, but you're right that it's pretty much the same thing. 'Oh, I know what the solution is so let's do XYZ' . Actually I've phrased that wrongly, it's usually more like 'You/we/I should do XYZ'. If asked why, responds with 'Because that's what's needed/the solution of course!'

I stand corrected about engineer, sir;)!

In particular, most of the people I've met who carry this title are wa-a-ay too prone to jump to solutions/decisions without preliminary analysis and study.

Yes - straight to implement solution.

At least that's what happens when you leave them to themselves. Very little "P" and a lot of "D". And then a lot of "R" ("redo"). :mad:

Ooh yes, very true. Though that particular behavioural pattern is not just limited to software engineers, either. Project managers of various kinds also do it a lot.

IMHO, the names and vocabulary do matter. Otherwise, it would be sufficient to call this particular method "1234" instead of "PD(whatever)A", wouldn't it? As far as everybody understands what all the four steps mean, why bother naming them, right? :)

I do not disagree. But would add that PDSA means more to some and PDCA means more to others. For others it could be their own accronymn or string of descriptives. The importance is the ability to rightly communicate the steps and their intent in a meaningful way.

And, by the way, I never heard anybody saying that Deming had revolutionized anything just by changing 'C' to 'S' in Stewhart's cycle. All I heard is that he improved understanding of the "Check" step and/or made it applicable more generally. IMHO, this hardly qualifies for a revolution, but it is a nice evolutionary step.


Whomever you heard this from must have been just another misinterpreter of both concepts.

I saw it in one of my primers along with it causing it to be a continuous cycle.

Overall I think you and I agree. Someone getting credit for small changes that, in my not as humble as it should be opinion, is one of my peeves. And the quality profession is full of people with flavor of the month which is just like the old vanilla but now it is called creamy vanilla. A lot of money can be spent on the same thing in a new package.

Your's free for just 3 easy payments of $29.95 + shipping. PDCA or PDSA it is the same to me. i know the tools, steps and process call it what you want. It is the Shewhart Cycle.

Mark

Overall I think you and I agree.


Yup. :agree:

No, but you're right that it's pretty much the same thing.
...
I stand corrected about engineer, sir;)!


:yes:

I have to get totally clean: my educational background is physics (theoretical physics = lots of math), so... I try to suppress, whenever possible, my tendency to nit-pick on things, but... you know, this background mix sometimes just can't be suppressed... :frust:



Ooh yes, very true. Though that particular behavioural pattern is not just limited to software engineers, either. Project managers of various kinds also do it a lot.


And now take those impatient managers and make them manage those impatient software "engineers"... now you know why most of the shrink-wrap commercial software around us has such a low quality! :notme:

Deming always credited Shewhart for the PDCA cycle. In Deming's book The New Economics he said that the PDSA cycle originated in his teachings in Japan in 1950 and appeared in Elementary Principles of the Statistical Control of Quality that year. Here are his definitions of each step:

Plan - somebody has an idea for improvement and develops suggestions that need to be tested

Do - carry out the test, comparison or experiment

Study - study the results and determine if met expectations and, if not, determine what went wrong

Act - adopt the change, abandon it, or run through the cycle again with different conditions, materials, people, rules

He always emphasized that to adopt the change or to abandon it requires prediction.

Bill Pflanz

Here are his [Deming's] definitions of each step:

Plan - somebody has an idea for improvement and develops suggestions that need to be tested

Do - carry out the test, comparison or experiment

Study - study the results and determine if met expectations and, if not, determine what went wrong

Act - adopt the change, abandon it, or run through the cycle again with different conditions, materials, people, rules

He always emphasized that to adopt the change or to abandon it requires prediction.


Here we go! Finally the most clear and unambiguous definition I've seen so far! (And it does mention the adopt/reject decision on the last step!)

Bill, excuse my ignorance: was this a precise quote of Deming's work?

Here we go! Finally the most clear and unambiguous definition I've seen so far! (And it does mention the adopt/reject decision on the last step!)

Bill, excuse my ignorance: was this a precise quote of Deming's work?

Yarik,

We should never stray too far from the Master's legendary work and teachings... Get "Out of the Crisis" and "New Economics...." True Masterpieces indeed and as fresh and applicable today as they were in the early eighties. Check Amazon.com, you may find some used copies...

Stijloor.

Here we go! Finally the most clear and unambiguous definition I've seen so far! (And it does mention the adopt/reject decision on the last step!)

Bill, excuse my ignorance: was this a precise quote of Deming's work?

Yes, it was copied directly from Deming's book on pages 134-136 only I left out the additional comments and examples. You can feel safe in letting your instructor know the definitions and source.

I don't know if it the book is still in print but the publisher was MIT.

Bill Pflanz

<snip> I don't know if it the book is still in print but the publisher was MIT.

Bill Pflanz

Bill,

They are still available. Look here (http://www.amazon.com/s/ref=nb_ss_gw?url=search-alias%3Daps&field-keywords=Deming).

Stijloor.

We should never stray too far from the Master's legendary work and teachings...

:yes::yes::yes:

...and we should never try to clutter them and distort their essense in our own derivative works...

Amen!


Get "Out of the Crisis" and "New Economics...." True Masterpieces indeed and as fresh and applicable today as they were in the early eighties.

According to yet-to-be-documented (and obviously yet-to-be-improved) BMS in our company, it will take me some time to go through "purchasing procedure" to get those books from Amazon. I've already started it, of course. :)

In the meantime, I'd greatly appreciate it if someone educated me: was the PDSA definition posted by Bill a literal quote of Deming's or just another interpretation?

:yes::yes::yes:

...and we should never try to clutter them and distort their essense in our own derivative works...

Amen!

According to yet-to-be-documented (and obviously yet-to-be-improved) BMS in our company, it will take me some time to go through "purchasing procedure" to get those books from Amazon. I've already started it, of course. :)

In the meantime, I'd greatly appreciate it if someone educated me: was the PDSA definition posted by Bill a literal quote of Deming's or just another interpretation?

Bill provided you with a (page) reference.

Stijloor.