Beyond PDCA by Praveen Gupta

[jrubio]

Before PDCA by Javier Rubio.

I found the article, and will attach it to judge for yourself what he is describing. Hope this helps

Before PDCA by Javier Rubio:

There are two wheels:

First: SDCA. Standardise Do, Check, Act (Standarization wheel).

Innovation

Second. PDCA. Improvement wheel. (Improvement Wheel)

When something do not work:

i.e:
Workers: Do they know the standards?.

Supervisors: Do standards exist?.

Worker...Supervisors....Coments..

Yes.........Yes........Put the standard in doubt.

Yes........No........Is the standard done?

No......Yes........Trainning and information required.

No....No.....The standard must be implemented by supervisors using SDCA.

 

[Steve Prevette]

Steve,

I can understand your concerns, but what do you think of Gupta's 4-P Model of Process Management? Do you throw it out because of his misconstrued interpretations / descriptions of the Quality Gurus' positions?

Doug

P.S. I am enjoying the discussions, but please do not leave it for ASQ - I am not a member. THANKS!

A fair question. As I look at the article, I see Figure 5. The decision to either ship it or improve is based solely upon if the item meets the specifications ("on target, acceptable"). If I meet specs, even though I may be outside of the UCL or LCL, I ship it and do nothing. Only if I fail to meet specs do I "Study and Reduce Variability". Only working on the "defects" is a sure way to fail at continual improvement.

Too much of the 4P's is based upon comparison to the target. He seems to think the PDCA/PDSA has you work to the specification limits (nothing could be farther from the truth) and his "big improvement" is to have you work to the target (apparently the center of the specification limits). I suspect the author has confused control limits (from SPC) with the specification limits (from the customer).

 

[Steve Prevette]

By the way, if you'd like to see the cycle I propose (and use at Hanford) please see http://www.hanford.gov/safety/vpp/piflow.htm

The PDSA cycle is embedded within this.

 

[jrubio]

By the way, if you'd like to see the cycle I propose (and use at Hanford) please see http://www.hanford.gov/safety/vpp/piflow.htm

The PDSA cycle is embedded within this.

Sometimes the proces is wrong by itself "per se" that why re-engineering is required. That is create a new process, not improve the wrong exist one.

Other impotant milestone is: Identify process and part statical variables to meassure.

 

[Steve Prevette]

Sometimes the proces is wrong by itself "per se" that why reingenieering is required. That is create a new process, not improve the wrong exist one.

Note that in the box "Determine Process Change Needed and Implement" perhaps the process change is a complete replacement of the process . . .

 

[vanputten]

Here is the text from ISO 9001. I hope I am not violating copyright. Figure 1 in ISO 9001 is a cyclical model of PDCA and includes the names of the PDCA clauses from ISO 9001. Clause 8 is the "Check" clause. I have provided the general requirements of clause 8 below.

Clause 5 - Management Responsibility (Act)
Clasue 6 - Resource Management (Plan)
Clause 7 - Product Realization (Do)
Clause 8 - Measurement, analysis, and improvement (Check)

The ISO 9001 text:
The model of a process-based quality management system shown in Figure 1 illustrates the process linkages presented in clauses 4 to 8. This illustration shows that customers play a significant role in defining requirements as
inputs. Monitoring of customer satisfaction requires the evaluation of information relating to customer perception as to whether the organization has met the customer requirements. The model shown in Figure 1 covers all the requirements of this International Standard, but does not show processes at a detailed level.

NOTE In addition, the methodology known as “Plan-Do-Check-Act” (PDCA) can be applied to all processes. PDCA can be briefly described as follows.

Plan: establish the objectives and processes necessary to deliver results in accordance with customer requirements and the organization's policies.
Do: implement the processes.
Check: monitor and measure processes and product against policies, objectives and requirements for the product and report
the results.
Act: take actions to continually improve process performance.

8 Measurement, analysis and improvement
8.1 General
The organization shall plan and implement the monitoring, measurement, analysis and improvement processes needed
a) to demonstrate conformity of the product,
b) to ensure conformity of the quality management system, and
c) to continually improve the effectiveness of the quality management system.
This shall include determination of applicable methods, including statistical techniques, and the extent of their use.

 

[jrubio]

Note that in the box "Determine Process Change Needed and Implement" perhaps the process change is a complete replacement of the process . . .

Dear Steve
I would include in your great diagram the sentence:

"process re-engineering "

and

Stablish a control plan & FMA (RPN)

 

[qualitytrec]

...The decision to either ship it or improve is based solely upon if the item meets the specifications ("on target, acceptable"). If I meet specs, even though I may be outside of the UCL or LCL, I ship it and do nothing. Only if I fail to meet specs do I "Study and Reduce Variability". Only working on the "defects" is a sure way to fail at continual improvement....I suspect the author has confused control limits (from SPC) with the specification limits (from the customer).

Steve,
I think I am misunderstanding your point possibly. Since business is about selling "product" then if product is in spec shouldn't it be shipped? Obviously if it is out of control then it should be reacted to and corrected, which I think is you point but you are not saying to reject product that is in spec because it is out of control limits are you?

Mark

 

[Steve Prevette]

Steve,
I think I am misunderstanding your point possibly. Since business is about selling "product" then if product is in spec shouldn't it be shipped? Obviously if it is out of control then it should be reacted to and corrected, which I think is you point but you are not saying to reject product that is in spec because it is out of control limits are you?

Mark

The issue is that the flow chart simply says to ship the product if it meets specs. But let's give this scenario - let us assume that the statistical control limits are well within the specification limits. What if you have a part that has exceeded the control limits, but is still within specification limits? The article's flowchart has you ship the product, and take no other actions. You don't do any troubleshooting until the specification limits are exceeded, and by then it is too late - you have made scrap metal.

Some companies might not even ship a product which exceeded statistical control limits - likely Toyota would not use the part. But even if you choose to ship the product, if there is a statistical signal that something has changed, then it should be investigated, not ignored.

 

[Jim Wynne]

The issue is that the flow chart simply says to ship the product if it meets specs. But let's give this scenario - let us assume that the statistical control limits are well within the specification limits. What if you have a part that has exceeded the control limits, but is still within specification limits? The article's flowchart has you ship the product, and take no other actions. You don't do any troubleshooting until the specification limits are exceeded, and by then it is too late - you have made scrap metal.

Some companies might not even ship a product which exceeded statistical control limits - likely Toyota would not use the part. But even if you choose to ship the product, if there is a statistical signal that something has changed, then it should be investigated, not ignored.

Yes Personally, I think that not shipping conforming product is foolhardy in most instances, unless an entire lot is involved, and there's good reason to believe--based on the statistics--that there's nonconforming product in there somewhere. Nevertheless, your point is well taken with regard to the strawman the article's author is using as an example. It's similar to the mythical Six-Sigma 1.5 sigma shift, which presupposes that operators won't take action until an out-of-control condition is encountered. In the real world, people don't go to the trouble of constructing charts and then ignore them, just as operators will often make corrections before statistical control is lost.