Quality Thinkers

Recent infrastructure failures and product recalls demonstrate — yet again — the importance of quality control procedures in manufacturing.


Posted on Sep 18, 2007

In this day of failing bridges, contaminated food, lead-bearing toys, and recalled cars, quality control must become a greater focus in our manufacturing and supply processes. The history of quality control bears re-examination. Key individuals working in the last century gave us the basis for statistical quality control and, thus, the means for modern manufacturing. The chain of reason flowed from Sir Ronald Fisher and Walter Shewhart, to W. Edwards Deming, to Japanese manufacturers, which put theory into practice to great advantage. Fisher (1890-1962), the English mathematician, is often credited with founding the modern study of statistics. His seminal work on process variance set the stage for later industrial quality control. Fisher's analysis of variance (ANOVA) is designed for small samples. Today, this method is used to pinpoint the causes of variation in manufacturing processes. Fisher redefined the purpose of statistics as the reduction of data, and he identified three core problems: the specification of the population that the data comes from, the estimation of the data, and the distribution of the data. Shewhart (1891-1967), an American mathematician, was a friend and colleague of Fisher's. He applied Fisher's basic concepts to removing variance within the manufacturing process and thus controlling quality. Shewhart created the archetypal control chart, which consists simply of three parallel lines on a flowing tape. The mid-line is the quality measure specified, and the lines above and below define the upper and lower limits of allowable quality. Variations in quality can be attributed to either an assignable or chance cause. The point of quality control work is to isolate and correct the assignable causes. Deming (1900-1993), brought up in the American Midwest, trained as a statistician but is best known as a teacher and a philosopher of business and manufacturing. He studied with Fisher in London and later worked with him in the United States. Deming also worked with Shewhart at Bell Labs. Whereas Fisher's work remained essentially academic, Shewhart had joined Western Electric and then Bell Labs and applied his quality analysis theories to business and engineering problems. Deming studied Shewhart's theory of statistical quality control and, in time, made Shewhart's rather inaccessible writing clear through his own books and lectures. Both Shewhart and Deming were influenced by Italian economist Vilfredo Pareto (1848-1923), who declared that 80% of the wealth of Italy was owned by 20% of the people, which became the famous 80/20 rule. The rule states that 80% of a problem comes from 20% of the causes, or the "vital few and trivial many." A contemporary of Pareto's, Joseph Juran (1904 -), also a statistician and manufacturing consultant, focused on managing for quality. He revised the 80/20 rule to mean a "vital few and useful many," indicating that the remaining 80% of causes of a problem shouldn't be ignored. All of these concepts have contributed to the statistical information that drives today's manufacturing, logistics, and supply chain, and more than ever should be the basis for the tools we use to improve our lives — and our bridges.


Top Enterprise Software Planning (ERP) Comparison