SBS - The best value in QMS software

SPC presentation - How to convince someone to use SPC - Grind diameters

C

CMfgT

#1
I get to give a presentation on Monday about implementing SPC into our job shop facility. Currently, we collect data (Statistical Production Control) for our customers. I make control charts and capability studies to evaluate the run after it has been complete. Now, I want to go towards what SPC is intended for Statistical Process Control. I am planning on showing all of our part numbers with key features, and they are mostly grind diameters and show stable and unstable processes along with good cpk values and bad cpk values and the corresponding defects per million. I am struggling with how to convince someone to use SPC that knows little about SPC. My big reason is this keeps operators from over reacting to over adjusting a machine. Any other suggestions, I guess there is just a ton of information and I don’t know where to go with it all. I know teaching operators to make/use a control chart takes time and collecting data takes time, but I want to show value, not non-value added activities.

Thanks
 
Elsmar Forum Sponsor

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#2
Re: SPC presentation

You can find a lot of information on X-bar and R charting, etc. But, unfortunately, it is not the correct SPC for grinding. I will forward to you training for the uniform distribution and X hi/lo - R chart. I hope that will help. The worst thing you can do is train people running grinders X-bar-R charting. :cool:
 
Last edited:

Steve Prevette

Deming Disciple
Staff member
Super Moderator
#3
Re: SPC presentation

Both Dr. Deming's Funnel Experiment (the issue with over-adjusting) and the Red Beads (convincer for SPC) work extremely well to overcome resistance.

I've got some SPC presentations on the internet at http://www.hanford.gov/rl/?page=1156&parent=1144

You are welcome to make use of them, they were developed under US government contract so there is no copyright. I do ask though if you use them to give attribution to myself and Fluor corporation.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#4
A nice overview of the techniques Steve mentions are in this link:

http://www.spcforexcel.com/ezine/nov2006_2/nov_2006pf.htm

However, be careful when interpreting the results of the funnel experiment with grinder operators. They need to know that they are not going to get the distribution that you get in that experiment - their machines will generate a uniform distribution (rectangle). The moral of its story is not that you have to leave the process set at one setting, but that you must not adjust the process until the control limits tell you it needs adjusted. If you can not make that clear after doing the funnel experiment (because of the difference between the distributions), then you may opt to drop it.

Here is the story I tell to precision machining people to get the point across:

I was sitting in a process planning meeting (APQP) when the engineers around the table proclaimed that the grinder they had would not be capable of running a part to a specific diameter tolerance. I felt they were just shooting from the hip, and I was curious if it was true.

I went out to the grinder and spoke to the operator. He was doing SPC on his operation, plotting the outer diameters. I looked at the chart, and it was a classic normal control chart with points randomly jumping about the mean.

I asked how often he was plotting his data, he said every two hours, just like the control plan said. I asked him how often he was adjusting his process, he said every 15 minutes.

My head dropped in dismay...

I asked him to try something different. I asked him to adjust his grinder to the lower control limit. I told him to ignore the mean. Run the grinder, and do not adjust it until the diameter reached the upper control limit. Then, adjust it back to the lower control limit.

He did that. Do you know how long it took to reach the upper control limit?

A week.

So, clearly his adjusting every 15 minutes to try to keep the machine at the mean was overadjustment. In fact, the operator had become the process. That made the process "normal", and most operator processes are normal distributions. But the machine process was not. It was a uniform distribution. It was masked by the unnecessary adjustments to the mean by the operator. CNC operators are notorious for overadjustment, because it is easy to push the buttons for an offset. I tell them if they want to push buttons, push the buttons on their calculator, not the machine.

Many quality professionals are fooled by seeing these supposedly 'normal' processes and their accompanying charts, and believe they really are. They use these charts to justify their claim that the process is indeed normal, in control and capable. Fools gold, my friends. It is usually garbage data.

X-bar-R charts encourage adjusting to the mean - and therefore encourage overadjustment in precision machining. That is one reason why they are the absolute worst chart for precision machining. For the uniform distribution, the mean has no meaning!

If I walk up to a precision machining process and see an X-bar-R chart exhibiting random variation about the mean, my first assumption is the process is out of control!

More often than not, I am correct.:cool:

BTW, CMfgT, that is the text for "Story No. 1" in the presentation I sent to you.
 
Last edited:

Stijloor

Staff member
Super Moderator
#5
A nice overview of the techniques Steve mentions are in this link:

http://www.spcforexcel.com/ezine/nov2006_2/nov_2006pf.htm

However, be careful when interpreting the results of the funnel experiment with grinder operators. They need to know that they are not going to get the distribution that you get in that experiment - their machines will generate a uniform distribution (rectangle). The moral of its story is not that you have to leave the process set at one setting, but that you must not adjust the process until the control limits tell you it needs adjusted. If you can not make that clear after doing the funnel experiment (because of the difference between the distributions), then you may opt to drop it.

Here is the story I tell to precision machining people to get the point across:

I was sitting in a process planning meeting (APQP) when the engineers around the table proclaimed that the grinder they had would not be capable of running a part to a specific diameter tolerance. I felt they were just shooting from the hip, and I was curious if it was true.

I went out to the grinder and spoke to the operator. He was doing SPC on his operation, plotting the outer diameters. I looked at the chart, and it was a classic normal control chart with points randomly jumping about the mean.

I asked how often he was plotting his data, he said every two hours, just like the control plan said. I asked him how often he was adjusting his process, he said every 15 minutes.

My head dropped in dismay...

I asked him to try something different. I asked him to adjust his grinder to the lower control limit. I told him to ignore the mean. Run the grinder, and do not adjust it until the diameter reached the upper control limit. Then, adjust it back to the lower control limit.

He did that. Do you know how long it took to reach the upper control limit?

A week.

So, clearly his adjusting every 15 minutes to try to keep the machine at the mean was overadjustment. In fact, the operator had become the process. That made the process "normal", and most operator processes are normal distributions. But the machine process was not. It was a uniform distribution. It was masked by the unnecessary adjustments to the mean by the operator. CNC operators are notorious for overadjustment, because it is easy to push the buttons for an offset. I tell them if they want to push buttons, push the buttons on their calculator, not the machine.

Many quality professionals are fooled by seeing these supposedly 'normal' processes and their accompanying charts, and believe they really are. They use these charts to justify their claim that the process is indeed normal, in control and capable. Fools gold, my friends. It is usually garbage data.

X-bar-R charts encourage adjusting to the mean - and therefore encourage overadjustment in precision machining. That is one reason why they are the absolute worst chart for precision machining. For the uniform distribution, the mean has no meaning!

If I walk up to a precision machining process and see an X-bar-R chart exhibiting random variation about the mean, my first assumption is the process is out of control!

More often than not, I am correct.:cool:

BTW, CMfgT, that is the text for "Story No. 1" in the presentation I sent to you.
Bob, please help me with this if you would.....

How do the following factors come in to play in terms of introducing variation, "control" and possible adjustments?
  • Dressing the grinding wheel
  • Machine wear
  • Coolant temperature
  • Hardness of material
  • The set-up of the grinder (in-feed, pressure between centers, etc.)
I appreciate your comments.

Thanks.

Stijloor.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#6
Bob, please help me with this if you would.....

How do the following factors come in to play in terms of introducing variation, "control" and possible adjustments?
  • Dressing the grinding wheel
  • Machine wear
  • Coolant temperature
  • Hardness of material
  • The set-up of the grinder (in-feed, pressure between centers, etc.)
Good questions. Let's go through them.

Dressing the grinding wheel
The X hi/lo chart tells an operator when to adjust a grinder based on wear. The need to dress the grinding wheel is can originate from two key factors - surface finish and key dimensional degradation. Trying to SPC surface finish is tough, although I suppose you can do it. Most grinding operators can visually recognize a finish that is starting to degrade and requires dressing. For key dimensional degradation, one example I have used was a part that had very complex compound curves. In order to maintain the profile, we had an optical comparator with a profile chart set at 75% of the profile tolerance zone. When the profile started to encroach that zone, we dressed the wheel. The sharpest radius (or corner, if you have one) is usually the first to go and the easiest one to track. We would adjust based on a simple diameter, dress on the profile degradation. We made some great parts doing that with an extremely low scrap rate.

Machine wear
First of all, precision machining is defined by all variation has been reduced to be statistically insignificant. So, the machine must be in sufficient repair so that the participation in variation from the equipment itself is insignificant. That being said, you will generally see the effects of a machine in disrepair in the R chart. Remember, in the X hi/lo-R chart, the R is roundness (or parallelism if a length rather than a diameter). When a grinder starts to go bad, or even CNC machines, the roundness starts to increase. It will guide to change bearings, worn chucks, etc. Usually an increase in roundness is a leading indicator that the tool needs changed for a CNC. But, if the roundness is still too high after changing the tool, then the mechanicals should be looked at until the roundness is brought back into control. By the way, you will not get any of this information from an X-bar -R chart.

Coolant temperature, hardness of material, and type of wheel grade
In X-bar-R charts, continuous improvement is considered when you compress the control limits closer to the mean. That is clear evidence that is is designed for a normal distribution - get the process closer to the mean. Uniform distributions are very different. The mean is meaningless. Compressing the limits just makes you adjust more frequently, creating overadjustment. So, how do you improve a uniform distribution? You work to make the slope shallower. The shallower the slope, the less operator intervention. Great! So, if you see the slope steepen, you may have harder material wearing away the wheel faster. You will have to adjust more often for harder material. If you change the grade of wheel, how do you know it is a better wheel? Usually, it is by the number of donuts the salesman brings in. But, the observant student would look for a shallower slope on the X hi/lo chart. Anything that affects the wear rate - or the slope - you can use the slope to determine improvement. The wear rate has no influence on the control limits - the X hi/lo limits are always set at 75% of the tolerance.

The set-up of the grinder (in-feed, pressure between centers, etc.)
The set up can affect the slope and the roundness. Too much pressure between centers will affect the roundness, maybe even cause chatter. It may also cause excessive wheel wear, increasing the slope. Too little pressure really hikes up the roundness! I would say that the roundness would be the key thing to watch for to ensure satisfactory setup.

What is amazing about all this? You will get meaningful data utilizing the X hi/lo-R chart for most facets of grinding control, especially the day to day adjustment for diameter. The chart is easily understood by operators - it makes sense. The math is simple, yet elegant. You will get incorrect signals from an X-bar-R chart, and the statistics utilized by that charting methodology is nonsense for precision machining. XI-R is also insufficient for precision machining, for many of the same reasons.

One of these days Marc is going to make me put this all in one thread...it's probably easier to just buy the book. :cool:
 
Thread starter Similar threads Forum Replies Date
S SPC Presentation and Explanation Training Materials for Beginners Training - Internal, External, Online and Distance Learning 2
Steve Prevette Presentation on SPC and Leading Indicators Statistical Analysis Tools, Techniques and SPC 3
bobdoering SPC for Precision Machining Presentation March 25, 2010 ASQ Central Kentucky Section ASQ - American Society for Quality 1
Steve Prevette A recent presentation on SPC and "Leading Indicators" Statistical Analysis Tools, Techniques and SPC 23
Z SPC Data presentation - The best way to present histograms in a dashboard format Statistical Analysis Tools, Techniques and SPC 14
bobdoering SPC for Precision Machining - ASQ Presentation Imported Legacy Blogs 0
W SPC Software - Graphical Presentation & p-chart LCL formula Statistical Analysis Tools, Techniques and SPC 4
A SPC procedure / flow chart Statistical Analysis Tools, Techniques and SPC 2
P Informational SPC in Batch Processing Statistical Analysis Tools, Techniques and SPC 3
B Handling lower detection limits for SPC and process performance Statistical Analysis Tools, Techniques and SPC 1
A How to reduce the process SPC monitoring Capability, Accuracy and Stability - Processes, Machines, etc. 3
P SPC for machining processes Capability, Accuracy and Stability - Processes, Machines, etc. 1
earl62 What is the best way to control special characteristics in Control plan? Is it Mandatory to have SPC for IATF 16949? IATF 16949 - Automotive Quality Systems Standard 12
J SPC - Stamping parts - Small batch size Statistical Analysis Tools, Techniques and SPC 6
T Attributes SPC study - Attributive control (Go gage) Statistical Analysis Tools, Techniques and SPC 5
S Anyone Familiar with "SPC for Excel" add-in? Lean in Manufacturing and Service Industries 4
G Assignable cause/corrective action list for SPC Software Statistical Analysis Tools, Techniques and SPC 3
H SPC for special characteristics - CSR - Control of SC and CC Statistical Analysis Tools, Techniques and SPC 1
J Determining SPC tolerance Statistical Analysis Tools, Techniques and SPC 21
L SPC - Methods to collect data IATF 16949 - Automotive Quality Systems Standard 7
Marc Steve Prevette's Statistical Process Control (SPC) "Library" Statistical Analysis Tools, Techniques and SPC 0
A SPC for Eddy Current Testing Statistical Analysis Tools, Techniques and SPC 28
S SPC (Statistical Process Control) for Unilateral Tolerance - Questions Statistical Analysis Tools, Techniques and SPC 6
Proud Liberal Informational Use of Classical SPC in Precision Machining Statistical Analysis Tools, Techniques and SPC 9
J Recommendations on Inspection/SPC Software Statistical Analysis Tools, Techniques and SPC 4
Q Real Time Electronic SPC/Control Charting Statistical Analysis Tools, Techniques and SPC 4
Rameshwar25 Other types of Control Charts described in Chapter II of SPC Manual Statistical Analysis Tools, Techniques and SPC 6
A Electric terminal crimping machines - Is SPC appropriate for this process? Manufacturing and Related Processes 4
J Properly understanding SPC - Newbie SPC questions Statistical Analysis Tools, Techniques and SPC 34
bobdoering SPC Charting frequency calculator for X hi/lo chart Imported Legacy Blogs 0
bobdoering Handy training tip for SPC - the Check Engine Light Analogy Statistical Analysis Tools, Techniques and SPC 0
QMMike SPC Software for MAC computers Statistical Analysis Tools, Techniques and SPC 2
Marc SPC based Dashboards - By Steve Prevette (November 2017) Statistical Analysis Tools, Techniques and SPC 1
H SPC Work Instructions (WI) Statistical Analysis Tools, Techniques and SPC 2
S Doubts about SPC taken in Machining - Part has +-0.01 Tolerance Statistical Analysis Tools, Techniques and SPC 5
A Short Term vs. Long Term SPC Study - Where is Cp and Cpk Defined Statistical Analysis Tools, Techniques and SPC 8
E SPC for Dummies - The correct meaning of all the graphs Statistical Analysis Tools, Techniques and SPC 3
Tagin Can/should SPC be applied to Computer Assembly and Software Imaging? Statistical Analysis Tools, Techniques and SPC 8
P Recommendations on SPC Software or Excel SPC add-in(s) Quality Assurance and Compliance Software Tools and Solutions 6
A Not all characteristics follow a Normal Distribution - How do you do SPC Chart Capability, Accuracy and Stability - Processes, Machines, etc. 5
T SPC in Insurance Back Office Operations Service Industry Specific Topics 4
C SPC for Particulate Alert/Action Levels Statistical Analysis Tools, Techniques and SPC 3
E SPC Production - Getting Cpk and Ppk Statistical Analysis Tools, Techniques and SPC 8
M SPC with many data points from one dimension Statistical Analysis Tools, Techniques and SPC 16
Q Mixed data collection (SPC) for different mold cavities Statistical Analysis Tools, Techniques and SPC 2
M How to Start SPC from the ground up! Statistical Analysis Tools, Techniques and SPC 8
bobdoering Available SPC and Metrology Consultants Paid Consulting, Training and Services 0
I SPC on Cyclic Data Statistical Analysis Tools, Techniques and SPC 10
S SPC Card for One Sample Size - Customer requires for all critical dimensions Statistical Analysis Tools, Techniques and SPC 8
R Quality Management & SPC Manufacturing and Related Processes 1

Similar threads

Top Bottom