The Elsmar Cove Business Systems and Standards Discussion Forums What is "Modified Control Chart"
 Forum User Name Keep Me Logged In Password
 Register Photo Albums Blogs FAQ Social Groups Calendar Search Today's Posts Mark Forums Read

 Elsmar Cove Forum Visitor Notice(s)

# What is "Modified Control Chart"

 Elsmar XML RSS Feed Monitor the Elsmar Forum Sponsor Links Courtesy Quick Links Links that Cove visitors will find useful in your quest for knowledge: International Standards Bodies - World Wide Standards Bodies International Standards Organization - ISO Standards and Information Quality Digest

Post Number #1
22nd January 2010, 05:46 AM
 jihyunpark Total Posts: 13
What is "Modified Control Chart"

Dear All,

I am preparing an annual quality review for my company and I have some references prepared by a previous person. Unfortunately I can not understand what he/she means it this guidline as below:

"Solubility, Disintegration, Sulfur dioxide, and Residue on ignition:
As the specifications are apparent for Solubility, Disintegration, Sulfur dioxide, and Residue on ignition, a modified control chart can be employed.
v How to construct Modified Control Charts
Fill in data sheets with all the annual data (k) of capsules determined once per lot and collected.
Calculate mean value, moving range (Rs), and average of moving ranges as described in the step for the construction of X-Rs Charts.
Calculate control limits in accordance with the below formula.
USL = Upper Specification Limit apparently predetermined
UCL (Upper Control Limit) = USL – 1.88×
Draw horizontal line for the control limits on the X and Rs charts as described in the section 4.4.1 and additionally draw horizontal line for upper specification limit. Evaluation method is described in the section 4.4.4."

My question is what is this control chart? and please give me a template if you get a one

Post Number #2
24th January 2010, 09:12 PM
 jihyunpark Total Posts: 13
Re: What is "Modified Control Chart"

is there anyone can help me soon?
Post Number #3
24th January 2010, 09:19 PM
 Stijloor Total Posts: 15,191
Re: What is "Modified Control Chart"

Quote:
 In Reply to Parent Post by jihyunpark is there anyone can help me soon?
It is weekend. Many of our members are off line. Tomorrow (Monday) you may get responses.

Stijloor, Forum Moderator.
Post Number #4
24th January 2010, 09:32 PM
 harry Total Posts: 6,286
Re: What is "Modified Control Chart"

Quote:
 ② Calculate mean value, moving range (Rs), and average of moving ranges as described in the step for the construction of X-Rs Charts.
Isn't it stated here that it is for the construction of moving range chart?
Post Number #5
24th January 2010, 10:50 PM
 Miner Total Posts: 4,042
Re: What is "Modified Control Chart"

It is slightly different. The end result is similar to statistically calculated pre-control limits. That is, these control limits are derived from both the moving range and the specification limits. They are calculated outside in from the specifications instead of inside out from Xbar.
 Thank You to Miner for your informative Post and/or Attachment!
Post Number #6
25th January 2010, 10:16 AM
 Darius Total Posts: 551
Re: What is "Modified Control Chart"

Agree with Miner

Don Wheeler, show it (altho don't recommend it) on his books "Advanced topics in Statistical Process Control", in 1995 edition, page 346.

He said that Modified Control Limits does nothing to encorage what the process is able to achive. He wrote
Quote:
 The rationale .. is the thought that as long as the subgroup average is within these limits, the process will be producing conforming product.
The control limits equations are

UCL = USL - 3 Sigma(X) + 3 Sigma (X_Average)
LCL = USL + 3 Sigma(X) - 3 Sigma (X_Average)

Quote:
 UCL (Upper Control Limit) = USL – 1.88×
Seems odd, maybe was obtained from real data from your process and parametrized as a constant (1.88).
Post Number #7
25th January 2010, 10:38 AM
 bobdoering Total Posts: 4,036
Re: What is "Modified Control Chart"

We know that similar control limit methodology has been proposed for precision machining (not by me, because the variation calculated is from measurement error, not process variation). To really understand if this is an issue or not we would have to understand the underlying distributions for the Solubility, Disintegration, Sulfur dioxide, and Residue on ignition measures. I would guess they are normal distributions, but would need some evidence to verify that. There may be lot variation issues that skew the long-term distribution - variation that can not be eliminated or controlled. But, if the overall distribution is normal, then these control limits are likely not correct.

If I was to guess why this was implemented, they were probably developed because the ongoing variation of these measures were not represented by the samples used for the control limit calculations, and therefore the calculated control limits could not be met over time. In frustration, they likely implemented this approach to provide some sanity to a situation the control charting methodology was not well understood. Rubber stamping just wasn't working for them.
Post Number #8
25th January 2010, 11:07 AM
 David DeLong Total Posts: 461
Re: What is "Modified Control Chart"

I covered Modified Control Charts when I attended the SPC seminar put on by Ford in Detroit in 1982 and it is also included in the book Statistical Quality Control by Grant & Levenworth.

Let's say you are drilling a hole and the diameter is considered "vital" so you have an X bar and R chart on it. Each time you have a drill change, the average changes so your X bar and R chart certainly looks out of control since tool changes affect the result.

The Modified Control Chart only affects the averages charts and the control limits are developed from the upper spec and lower spec limits.

UCL modified = Upper Spec limit - (V X standard deviation estimate)
LCL modified = Lower Spec limit + (V X standard deviation estimate)

V is a constant that will be found in most statistical books and is 1.66 for a 5 part sample.

The estimate of the standard deviation is developed from the R chart and is the average range/ d2 (also a standard per sample size).

Modified Control Chart control limits are sometimes called "shut down limits" since if one found a point outside the limits, there is a probably of finding parts outside the specification limits and the process should be shut down. In other words, the control limits are developed to give one a Cpk of 1.00.

The automotive industry do not recognize this type of chart but with disposable tooling or if one had a process where there is a lot of wear, it may be the most appropriate choice.

Hope this helps.
 Thank You to David DeLong for your informative Post and/or Attachment!

 The Elsmar Cove Business Systems and Standards Discussion Forums What is "Modified Control Chart"

 Bookmarks

 Visitors Currently Viewing this Thread: 1 (0 Registered Visitors (Members) and 1 Unregistered Guest Visitors)

 Forum Posting Settings You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Emoticons are On [IMG] code is On HTML code is Off Forum Rules

 Similar Discussion Threads Discussion Thread Title Thread Starter Forum Replies Last Post or Poll Vote Jim Wynne Statistical Analysis Tools, Techniques and SPC 86 23rd September 2016 04:12 PM Izoyd Statistical Analysis Tools, Techniques and SPC 4 15th August 2014 08:00 AM astudent FMEA and Control Plans 19 19th May 2012 08:23 AM RCW ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 12 2nd March 2012 09:37 AM jihyunpark Statistical Analysis Tools, Techniques and SPC 1 26th February 2010 08:00 AM

The time now is 03:09 AM. All times are GMT -4.