Systems Thinking Is Always Relevant
Watergate’s Deep Throat—A Systems Thinker” by Clare Crawford-Mason (November 2004, p. 61) reminded me of a personal incident that took place in the mid to late ’80s when my job was introducing me to management and quality theory.
I stumbled across an undated copy of an article on systems thinking by Russell Ackoff, whom I didn’t realize was and still is one of the foremost experts on the subject. I located his university office and naively left a message saying I wanted to discuss the article with him. This prominent, extremely busy and important man returned my call a week or so later, humbly apologizing for the delay due to having broken his leg, if I remember correctly.
I told him his article was fascinating and thought Ford Motor Co., my employer at the time, could benefit greatly from his ideas. Not wanting to dampen my youthful enthusiasm, he gently mentioned, “What you have isn’t really new information. I wrote that article in the ’50s.” Like all groundbreaking concepts, systems thinking never loses its relevance, always seems fresh and re-emerges when needed.
We quality professionals use many tools that help us think systematically, but systems thinking is much more than that. Truly gifted people are systems thinkers, and they become historically significant when they have the courage to act on their convictions despite significant short-term risk. They understand what is best for a country is not necessarily what is best for a state, a special interest group or a political party.
Similarly, we must remind ourselves what is best for society or a business is not necessarily what is best for a department, an employee, an audit or our largest customer. To really understand a particular issue and initiate successful change, we must understand the issue’s context and the change’s impact on the larger system.
Get To Know a Company Before Merging
I read with interest Jonathan Lore’s article, “Five Lessons From a U.S.-English Merger” (November 2004,p. 30). It is a typical problem faced by many American businesses and illustrates our country’s approach to foreign policy as well.
The original consultant for Churchill Manufacturing had it right when he first tried to understand the operation. As a matter of fact, the CEO and CFO should have taken the time to get to know Churchill before entering into the merger.
As Americans, we see issues and problems through our eyes and based on our background. This often creates problems after the euphoria of the merger has worn off because we end up misunderstanding the environment in which our international partner operates. This lesson should be taken to heart by all companies involved with international operations.
Only One Person’s Opinion Matters
here has been much musing in ASQ publications about the future of quality, particularly as it relates to the role of the quality professional. “The Future of Quality: Customer Value” by Armand Feigenbaum and Donald Feigenbaum is the latest example (November 2004, p. 24). I compliment the authors on the content and logic of their article but cannot help feeling the tone of the piece illustrates precisely why the quality profession is wandering in the wilderness.
The authors position the alignment of management and quality to achieve customer value as a recent phenomenon. Herein lies the fundamental problem bedeviling the quality profession. Quality has always been about delivering value to an organization’s customers, but we have not fully engaged ourselves in this central component of the quality equation. “Value,” we think, “is some term used in marketing to disguise the low prices!”
Instead, we’ve focused our attention on the internal aspects of quality and surrounded them with arcane language and practices that have further alienated us from the business mainstream. We’ve flitted from one quality program to the next, always asking management to faithfully accept we will deliver. We’ve failed to understand the lackluster performance of our programs in delivering tangible and lasting financial improvements is the reason for our declining credibility.
We must understand only one person’s opinion matters in determining quality and value: the customer’s. We are trying to pry money from his or her wallet, and we have to constantly prove it is a wise decision to purchase our product. The minute we lose sight of this we marginalize our role.
Borcom Associated Consultants
Vancouver, British Columbia
Histograms Don’t Explain Time Sequence of Data
alerie Funk’s “Back to Basics” column in the November 2004 issue (“Early, Late or Right on Time?” p. 104) was a good overview of histograms and frequency distributions. However, the concluding sentence is erroneous. She says, “A histogram can give you a quantitative picture of a data set over a period of time.” If Quality Progress had cleared this article through the Statistics Division, certainly Davis Balestracci or another member would have pointed out the fallacy of the statement.
You see, a histogram tells you nothing about the time sequence of the data. Balestracci has written several papers that present examples of data with clear patterns and trends within the data when you look at them in time sequence, but collapse down to a nice looking bell shaped curve.
Even the rocket scientists at a major aerospace corporation once fell into this trap. They could not understand what was happening with a certain hole size being drilled for a holding bolt. The distributions were just fine, but the holes were frequently too large or too small in size. If you plotted the hole sizes in the time sequence in which they were drilled, the answer was obvious: tool wear. The holes were progressively getting smaller, but you couldn’t tell that from collapsing all the data into a histogram.
Fluor Hanford and City University
Editor’s Note: We apologize for the errors in the column and thank our readers for pointing them out to us.
Initial Use of P-Chart Fundamentally Incorrect
Forrest Breyfogle’s example showing how a p-chart applied to satellite view data will produce many special cause signals is correct, but I don’t think it’s because of a philosophical difference between W. Edwards Deming and Walter Shewhart (“3.4 per Million: Control Charting at the 30,000-Foot-Level, Part 2,” November 2004, p. 85).
His initial use of the p-chart for aggregated defect data is fundamentally incorrect. The p-chart’s upper and lower control limits depend on assuming the data can be modeled with a binomial distribution. The sum of many binomial distributions doesn’t equal another binomial distribution, unless their p-values are identical.
Breyfogle suggests using a
Pareto chart to stratify the defects, so it’s likely
there’s more than one defect category. Don Wheeler, in
Advanced Topics in Statistical Process Control,
pp. 257-259, concludes an XmR chart will give the user empirical limits that will perform satisfactorily.
JOHN J. O’NEILL JR.
Six Sigma Alliance
A New Way To Study For Certification Exams
I have four ASQ certifications and am planning to take the certified quality manager exam this spring (James J. Rooney, “Certification Exam Tips, Trips and Traps,” October 2004, p. 41). With my business schedule, my family’s schedule and other activities, I developed an effective way to study over the past several years. I read the manual and questions only once and take practice tests.
Plan: I order a study manual and answers and determine how many pages I need to cover each week to finish the manual in six to eight weeks. Whenever I read the manual, I read it out loud and record my readings on a tape.
Do: I play the recording anytime I’m driving in the four to six weeks prior to the test. I can usually recite the entire manual within a few weeks. While it’s not party conversation, it is very effective for test taking.
Check: I take the practice test three to four weeks before the final exam and give myself the same amount of time I will have during the actual exam.
Act: I write down any questions I was not sure of or missed altogether, including the explanation of each possible answer. I take one more timed exam one week before the final exam and again make note of the few questions I am unsure of. I flag any areas I’m still uncomfortable with in my study manual and refer to it during the test if needed.
This practice takes one to two hours each week for eight to 12 weeks before the exam and uses my daily hour of travel time to prepare.
Alcoa-Flexible Packaging Division
Project Managers Cannot Handle Quality
I was extremely disappointed by Russ Westcott’s article “The Metamorphosis of the Quality Professional” (October 2004, p. 22). With it and the recent career related articles by Greg Hutchins, I wonder why ASQ is not looking at disbanding.
Westcott says quality professionals will disappear in the next eight to 10 years. For the past five years, I have worked in organizations where activities that were traditionally performed by quality professionals were delegated to other parts of the organization. This strategy was unsuccessful when organizational change ensued.
One such organization is struggling with a way to approach TS 16949 because the operations personnel who were managing the quality management system were not prepared to change models. Another company is facing financial ruin because it was unable to address certain quality concerns due to a lack of expertise. The director in charge of quality was a manufacturing engineer by training and ill prepared to deal with customers’ quality issues.
To see the lead article in Quality Progress promoting the idea that project managers can handle quality is irresponsible. I also noticed both Westcott and Hutchins are consultants. I believe in the interest of fairness and a balanced presentation, Quality Progress should try to find an author who is working in a corporate or facility environment and get his or her view of the future of the quality professional.
The number of arrival times in the histogram in the November 2004 “Back to Basics” column (Valerie Funk, “Early, Late or Right on Time?” p. 104) exceeds the number of workdays tracked. Both numbers should be the same.