Q: One of our inspectors is not grasping the concept of rework verification. Can you provide a simple explanation of what rework verification is and what it isn’t?
A: Let’s start with the definitions of "verification" and "rework." According to ISO 9000:2005, verification is "confirmation through the provision of objective evidence that specified requirements have been fulfilled." The standard defines rework as "action on a nonconforming product to make it conform to the requirements."1
Armed with those definitions, explain to the inspector that verification of rework means confirming through objective evidence that action on a nonconforming product has been taken to make that product meet the requirements. From a common–sense point of view, rework verification is simply making sure rework meets the requirements through inspection, testing or both.
Most likely, this concept simply does not make sense to your inspector. Make sure you explain rework verification and why it is necessary in a way that makes sense to that individual.
It would help a great deal if you had a standard operating procedure (SOP) for rework verification. If you don’t, I suggest creating one and then asking that particular inspector to follow it. A good SOP will explain in a concise manner why the SOP is required in the first place.
If the previous suggestions do not work, you may be better off finding that individual another job within the company or letting him or her go. I don’t see how a company can have someone working in the area of quality management who can’t understand the concept of rework verification.
Mehta Consulting LLC
- International Organization for Standardization, ISO 9000:2005—Quality management systems—Fundamentals and vocabulary.
For more information
- Mehta, Pradip, "Repair vs. rework," Expert Answers, Quality
November 2009, p. 9.
Q: What can we do when top management is not involved enough in the organization’s quality system?
A: I would suggest you look at the focus of your quality system. If your quality system is going to be successful, it should be a tool management can use to evaluate the organization’s performance.
Top management needs to lead the organization if it’s going to follow through on its strategies and achieve organizational goals. A robust quality system gives management the information and tools it needs to make good decisions.
I would suggest evaluating your current quality metrics. Are they aligned with what your top management needs to lead effectively? For example, the Baldrige criteria link strategy to organizational metrics and require leaders to evaluate the organization’s performance.
Leaders need useful information to accomplish their work. You need to make sure you’ve linked your quality metrics to your organization’s strategies and that you’re providing the information they need. One way to accomplish this is to implement a balanced scorecard as part of your quality system.
There are multiple frameworks organizations can use to create a balanced scorecard and build a quality system around their strategies. The framework constructed by Robert Kaplan and David Norton, for example, uses these categories:
- Internal business process.
- Learning and growth.
To use this framework, organizations would need to identify strategies and goals in each of those areas and then identify metrics to monitor how the organization achieves results. Another approach by Quint Studer opts for a five-pillar approach involving service, quality, financial, people and growth.2
Both frameworks encourage organizations to identify strategies and metrics. By adopting one of these methods, you can link your quality system to management’s goals and provide it with useful tools to evaluate organizational performance. This will get top managers involved in the process and encourage them to support the organization’s quality system.
New Lisbon, WI
- Robert S. Kaplan and David P. Norton, The Balanced Scorecard: Translating Strategy Into Action, Harvard Business Press, 1996.
- Quint Studer, "The Five Pillars," www.studergroup.com/dotCMS/knowledgeAssetDetail?inode=109970 (case sensitive).
For more information
- Sherman, Peter J., and James G. Vono, "All Ears," Quality
July 2009, pp. 16–23.
Q: Literature I’ve read says "the same part is measured 10 times by three different operators using the same gage." Some people interpret this to mean the same physical part is measured at the same point on the same feature 10 times by three separate operators. Other people interpret this to mean that 10 specimens of the same part number are measured with the same gage on the same point on the same feature by three separate operators.
The confusion comes from the expression "measure the same part." Which is the correct interpretation: measure one part specimen 10 times on the same feature or 10 parts at the same feature? What if you want to perform gage repeatability and reproducibility (R&R) on a part, but only one is being manufactured (due to size or cost) and additional parts are not available?
A: Based on the way the literature you quoted is written, it is easy to understand why there’s confusion about the number of parts or the number of times you measure one part. This is why gage R&R is a tool for which I usually turn to reference books.
First, let’s address the number of parts. In their book, EMP: Evaluating the Measurement Process, Donald J. Wheeler and Richard W. Lyday write, "Five parts were used in order to introduce some product variation into the EMP study." They add, "One simple way to ensure the EMP study contains a reasonable amount of part-to-part variation is to select one sample part or batch on each of several successive days of operation."1 Based on this material, it appears that multiple parts are desired as opposed to multiple measurements of the same part.
In Implementing Six Sigma: Smarter Solutions Using Statistical Methods, Forrest W. Breyfogle III writes, "Five samples selected from a manufacturing process are to represent the normal spread of the process."2 Again, the implication is multiple parts rather than multiple measurements of the same part.
According to the first edition of The Certified Six Sigma Black Belt Primer from the Quality Council of Indiana, "Choose five parts at random and select a quality characteristic to measure."3 This seems to be consistent with the other two sources. So, based on a literature search, it appears the implication is for multiple parts rather than one part measured multiple times.
Regarding the second part of your question about gage R&R on one part, when you’re talking about measurement error, multiple components must be evaluated.
Part variability is not applicable for a situation with one part, but different operators would be a factor. You must look at the difference in measurement instruments. The feature being measured can lend itself to variability. And the time of day at which the measurement occurs should also be a consideration.
If this one part doesn’t go beyond an environment of one measuring area in one physical location being measured by one operator using a single measuring device, then the evaluation of variation error is very straightforward. Keep in mind that when you have only one part, it just means you have no part-to-part variability. The other sources of measurement error are still in play.
AVP Continuous Improvement
Lincoln Financial Group
- Donald J. Wheeler and Richard W. Lyday, EMP: Evaluating the Measurement Process, second edition, SPC Press, 1989, pp. 37, 40.
- Forrest W. Breyfogle III, Implementing Six Sigma: Smarter Solutions Using Statistical Methods, Wiley and Sons, 1999, p. 229.
- Quality Council of Indiana, The Certified Six Sigma Black Belt Primer, first edition, Section VI, p. 98.
For more information
- Ermer, Donald S., "Appraiser Variation in Gage R&R Measurement," Quality Progress, May 2006, pp. 75–78.