Quality Exclusives

Readers Respond: "Quality Control Should Check the Process, Not the Product"

April 16, 2012
/ Print / Reprints /
/ Text Size+
Is this statement truly feasible? Quality Magazine LinkedIn discussion group members debate the answer.

William Edwards Deming
Source: Quality Magazine

Quality Magazine’s LinkedIn Group members sound off about this statement and the nature of process improvement. Share your comments below or in our LinkedIn Group today.

Aziz Milad, Quality Manager at Franke Kitchen Systems Egypt:

“I like the statement by Dr. Deming: "Quality control should check the process not the product."

Bob Doering, Advanced Product Quality Planning Engineer at Stoneridge Hi-Stat Division:

”The goal is to control the process, not the product. So, rather than measuring product from the stream, the goal is to control the variables of the process: pressures, feeds, speeds, etc. Unfortunately, there is physical phenomena that cannot easily be measured directly, such as tool wear. You can measure the part, and use it as surrogate of process parameters, but you cannot measure the tool (always) directly. So, Deming's point of controlling the process-not the product, is admirable, [but]not always feasible.”

Kim Riehman, Quality Manager at Cast Aluminum Solutions:

“Bob, good example! Which is easier to do, measure the product size and maybe surface finish, or pull the tool (and maybe tool holder), somehow measure for tool wear and go from there? Oh, and also replace the tool-toolholder with another pre-qualified setup...I know which one I would pick.”

Tom Johnson, Environmental Health and Safety Director at Kirtland Products:

“Is the point that sometime product review can indication something is going wrong with the process? Something not otherwise detectable without shutting down the process?”

Rick Post , Lead Auditor at Thyssenkrupp Waupaca:

”Everything you do is a process... part goes into the press... gets pressed... is removed from the press... a process. I should think that a robust quality system would have a procedure for checking the tooling at regular intervals to ensure conforming parts... again a process. And if the tooling is found to be wearing to the detrimental side, don't you have a process for contacting the customer or generating a work order for replacement tooling or refurbishment? Another process... and at the least I should think that the operator has something... a gauge... a camera... lasers... something checking the parts in real-time to prevent worn tooling from becoming an issue... another process. If all these processes (and obviously others) are functioning properly you will not get a systemic non-conformity.”

Bob Doering:

“You have to look at Deming's statement in context. He was using the term in the more common generic "everything is a process" definition, as used in modern quality systems. He was looking to get away from looking at the product, and look at the things that generate the product and control their variables. And underlying point is control the variables in the processes that make the product to make good product consistently. One you are looking at the product, you are past the processes he was referring to.

“Once you shut down a process, you change it. As a matter of fact, start-up and warmup are "special causes" and are not typically stable! So, you need to get data from an ongoing, steady state process to get effective data.”

Richard Harpster Owner at Harpco Systems, Inc.:

“It would be interesting to know when Dr .Deming made the statement. I wonder if it was in reaction to him seeing people perform Statistical Product Inspection (SPI) rather the Statistical Process Control (SPC) which he was trying to teach people. It would be interesting to do a study of the SPC charts in plants to see home many measure a process characteristic (i.e. soaking bath temperature) and how many measure a product characteristic (material thickness). My guess is that the majority of the plants have implemented SPI. Thus you have Dr. Deming's comment.”

Kim Riehman:

“Richard, based on my experience I think you're absolutely correct. I also think that occurs because the customer demands statistical data on certain drawing features and says something like, "yes, your process data is nice, but show me how that translates to my parts on this feature or that...Most companies won't be able to close that loop.”

Tom Johnson:

“It does not seem that a consensus has been formed here. I will reiterate my initial assertion: Quality processes which ignores product puts the customer at risk. At some point there must be process validation, a product validation. This validation is a process, of course, but the product (or service) is the proof.

“Focusing on the process solely makes the inherent assumption that the process will provide the desired product or service. This can be a dangerous assumption.”

Bob Doering:

“I don't think there is an argument that quality processes should ignore the product. The assertion is not to ignore the product, but rather to place focus on the thing that creates the product, as that is origin of the variation. The product is simply the result, and as such, is ex post facto of the actual source of variation.”

Rick Post:

“The product should never be ignored... after all that is what the customer is actually paying for... but the point is: If you have mature processes with competent operators you should be able to ensure a quality product, plain and simple... one of those mature processes obviously needs to be some form of part/product inspection. That process should be done by the process owner, not quality.

“Look at this from a lean perspective... quality is basically a non-value added department; if it were responsible for inspecting all the product leaving your facility how large would that department have to be? Now look at your overhead and profits... where did they go? Into inspection... can you inspect quality into a product..? No...it must be designed into the process.”

Bob Doering:

“Quality is not "value added," but it is "value maintained," which makes it equal to value-added processes, and not overhead. That "not-value-added" concept is the biggest misconception out there, and why some companies fail when the cut quality first as simply overhead using that logic. Still, like "value added" steps, it needs to be well thought-out and effective to ensure profits.”

Aziz Milad:

Dearest colleagues: I have noticed that some are talking about process other about product, in fact both are the same as a piece of coin with two faces .....no one can imagine that a not properly maintained and controlled process (I mean by this documented, distributed, audited, awarded to workers and [training] them on it) such process can't deliver unsatisfactory product.

“By other means when we formulate the process by this means and put quality tools in place ......we guarantee good product. Sustainability is then maintained by sampling, testing, customer satisfaction and expectation measures.”

Todd F. Wilson, Heavy Civil General Contractor-Senior Project Engineer at Austin Bridge & Roadway:

“I understand completely. If you control the process you control the outcome thus the product. Managing the process is the key to success. I have never read about Deming only heard snippets about the man but I can tell you I know this principle to be true with or without him.”

Did you enjoy this article? Click here to subscribe to Quality Magazine. 

You must login or register in order to post a comment.

Proper process equals acceptable product

Clifford Slaven
April 18, 2012
I first heard of Dr. Deming from Dr. Jim L. Windle, Instructor at Purdue University. It really is common sense. If the product is bad, something went wrong during the process. When the process is right, the product will be right.

Mass Inspection v Process Control

Ralph Jackson
April 18, 2012
Quality of product is an attribute of value to customers. Quality Control is a departmant that does not create or maintain product quality, it only serves to evaluate product quality, to me this is the very definition of waste. I think Dr. Demming was pushing against the idea that ongoing, mass inspection is required.

Product or Process

B. Helenbart, Process Improvement
April 18, 2012
The Product is evidence of process effectiveness. It provides clues to where the process may be failing. However, it is much easier and less stressful to inspect the product. Improving processes involves people and politics with resistance to change. Product does not talk back. Most managers will construct Quality Controls about the product to keep their jobs.

"Quality Control Should Check the Process, Not the Product"

April 18, 2012
There is a deep meaning in Dr.Deming's statement. It also makes a relation with Dr.Ishikawa's Cause and effect relationship. when we see product as an effect, then all the process controls are its causes. It means, when you control your process variables, you are assuring that the product produced should meet the specifications. Hence there is a strong causal relationship between process and product.

Product or Process

Mike Epps
April 18, 2012
If I were to implement spc, obviously that would involve someone (i.e., the operator) to measure parts at some frequency. As the quality inspector or auditor part of my task would be to audit the spc control chart along with other process parameters. Essentially I would not have to touch the product at all. However, realistically as an inspector or auditor the joy of inspections is to observe the product even if it means implementing a sampling plan.

Quality Control

Dennis Craggs
April 18, 2012
Initially, quality was checked in the field. The product complaints from the customer was the metric. This prompts 100% quality inspection, but is not 100% efective. The more effective approach is to move further upstream in the product creation process. With SPC and control of process variables, quality improves. Further upstream, design for quality is even more effective. Then SPC is just a confirmation of the product and process design.

Product vs. Process

Chris Lee - Applications Manager - Corning Tropel Metrology Instruments
April 18, 2012
I think there is some misinterpretation of the intent of the statement. I have a perfect example of this type of control. My company used to co-market a product that measured the scatter from a surface caused by roughness. It was an instantaneous non-contact measurement device, and could easily be installed in any production line. The sales were very disappointing for this product. Why was this? There was not a direct profile measurement (no Rz no Rpk, nothing that was specifically on the product prints). The measurement correlated to a change in process, so it was inspecting the product for a CHANGE in the process. With this tool you could detect the change in process, pull a part off the line when detected and measure off-line for the parameters that were on the "product" print. This tool was a perfect example of controlling the process instead of the product. It does not eliminate the need to periodically inspect the product, but it allows the process to be controlled independently of the product specifications to detect the need to make corrections to the processing elements. At the very least it would greatly reduct the inspection load on the quality control lab.


Scott lebeter
April 18, 2012
Product validation is also a process. Each process is a variable which needs to be monitored to asses total process capability. Frequency and sample size may be minimized once process capability is understood.

Product or Process

Paul Young, Quality Mgr.
April 18, 2012
Both need to be checked. By whom and how frequently often depends on many variables such as company structure, the process itself, process automation, human impact, customer requirements. In my opinion the process owners should check the process and inspectors check the product.(Inspection supervision checks the inspection process as the owner of that process). Our customers require record keeping of product conformance, not process. If all you do is "wait" to check product compliance you will have high scrap, rework costs and higher customer rejections.

Flexibility is key

Don LaDue
April 19, 2012
In the real world there are no simple solutions. Product evaluation is key to identifying problems. Process control is key to eliminating them, and can be used eliminate unnecessary inspection time.

Taught by the good Dr. Deming: Validate or Verify

April 23, 2012
Not all processes lend themselves to the ideal, i.e., automation with controls that eliminate the need to verify. The need for Verification for some processes will never go away.


Don Davenport
June 22, 2013
Are we forgetting SPC here?




 In honor of World Quality Month, we spoke to James Rooney, ASQ Past Chairman of the Board of Directors 2013, for his take on quality around the world.
For more information, read the ASQ Speaking of Quality column.
More Podcasts


Quality Magazine


2014 April

Check out the April 2014 edition of Quality Magazine for features!
Table Of Contents Subscribe

Manufacturing Process

Has/does news about a manufacturers’ recall (like the GM recall in the news now) cause you or your company to reexamine its manufacturing process?
View Results Poll Archive

Clear Seas Research

qcast_ClearSeas_logo.gifWith access to over one million professionals and more than 60 industry-specific publications,Clear Seas Research offers relevant insights from those who know your industry best. Let us customize a market research solution that exceeds your marketing goals.



facebook_40.png twitter_40px.png  youtube_40px.pnglinkedin_40px.png