Walter A. Shewhart

Walter Andrew Shewhart (pronunciado como "Shu-jart", 18 de marzo, 1891 - 11 de marzo, 1967) fue un físico, ingeniero y estadístico estadunidense, a veces conocido como el padre del control estadístico de la calidad.

W. Edwards Deming dijo de él:

Como un estadístico, él era, como muchos de nosotros, autodidácta, con un buen background de física y matemática.

As a statistician, he was, like so many of the rest of us, self-taught, on a good background of physics and mathematics.

Early life y educación

Nacido en New Canton, Illinois de Anton y Esta Barney Shewhart, asistió a la University of Illinois antes de obtener un doctoraado en física de la University of California, Berkeley en 1917.

Trabajo sobre la calidad industrial

Bell Telephone’s engineers had been working to improve the reliability of their transmission systems. Because amplifiers y other equipment had to be buried underground, there was a business need to reduce the frequency of failures y repairs. When Dr. Shewhart joined the Western Electric Company Inspection Engineering Department at the Hawthorne Works en 1918, industrial quality was limited to inspecting finished products y removing defective items. That all changed on May 16, 1924. Dr. Shewhart's boss, George D Edwards, recalled: "Dr. Shewhart prepared a little memorandum only about a page en length. About a third of that page was given over to a simple diagram which we would all recognize today as a schematic control chart. That diagram, y the short text which preceded y followed it, set forth all of the essential principles y considerations which are involved en what we know today as process quality control."​ Shewhart's work pointed out the importance of reducing variation en a manufacturing process y the understanding that continual process-adjustment en reaction to non-conformance actually increased variation y degraded quality.

Shewhart framed the problem en términos de assignable-cause y chance-cause variation y introduced the control chart as a tool for distinguishing between the two. Shewhart stressed that bringing a production process into a state of statistical control, where there is only chance-cause variation, y keeping it en control, is necessary to predict future output y to manage a process economically. Dr. Shewhart created the basis for the control chart y the concept of a state of statistical control by carefully designed experiments. While Dr. Shewhart drew from pure mathematical statistical theories, he understood data from physical processes never produce a "normal distribution curve" (a Gaussian distribution, also commonly referred to as a "bell curve"). He discovered that observed variation en manufacturing data did not always behave the same way as data en nature (Brownian motion of particles). Dr. Shewhart concluded that while every process displays variation, some processes display controlled variation that is natural to the process, while others display uncontrolled variation that is not present en the process causal system at all times.​

Shewhart worked to advance the thinking at Bell Telephone Laboratories from their foundation en 1925 until his retirement en 1956, publishing a series of papers en the Bell System Technical Journal.

His work was summarised en his book Economic Control of Quality of Manufactured Product (1931).

Shewhart’s charts were adopted by the American Society for Testing y Materials (ASTM) en 1933 y advocated to improve production during World War II en American War Standards Z1.1-1941, Z1.2-1941 y Z1.3-1942.

Later work

From the late 1930s onwards, Shewhart's interests expanded out from industrial quality to wider concerns en science y statistical inference. The title of his second book Statistical Method from the Viewpoint of Quality Control (1939) asks the audacious question: What can statistical practice, y science en general, learn from the experience of industrial quality control?

Shewhart's approach to statistics was radically different from that of many of his contemporaries. He possessed a strong operationalist outlook, largely absorbed from the writings of pragmatist philosopher C. I. Lewis, y this influenced his statistical practice. en particular, he had read Lewis's Mind y the World Order many times. Though he lectured en England en 1932 under the sponsorship of Karl Pearson (another committed operationalist) his ideas attracted little enthusiasm within the English statistical tradition. The British Standards nominally based on his work, en fact, diverge on serious philosophical y methodological issues from his practice.

His more conventional work led him to formulate the statistical idea of tolerance intervals y to propose his data presentation rules, which are listed below:

1. Data has no meaning apart from its context.
2. Data contains both signal y noise. To be able to extract information, one must separate the signal from the noise within the data.

Walter Shewhart visited India en 1947-48 under the sponsorship of P. C. Mahalanobis of the Indian Statistical Institute. Shewhart toured the country, held conferences y stimulated interest en statistical quality control among Indian industrialists.​

He died at Troy Hills, New Jersey en 1967.

Influencia

En 1938 su obra llama la atención de los físicos W. Edwards Deming y Raymond T. Birge. Ambos estaban profundamente intrigados por el issue de la medición del error en ciencia y habián publicado un paper landmark en Reviews of Modern Physics en 1934. Al leer los insights de Shewhart, escribieron al journal para recast totalmente su approach en los términos de lo que Shewhart advocated.

The encounter comenzó una larga colaboración entre Shewhart y Deming que incluyó trabajo sobre la productivity durante la Segunda Guerra Mundial y Deming's championing de las ideas de Shewhart en Japón desde 1950 en adelante. Deming desarrolló algunas de las propuestas metodológicas de Shewhart acerca de la inferencia científica y llamó a su síntesis el ciclo de Shewhart (Shewhart cycle).

Logros y honores

En su obituario para la American Statistical Association, Deming escribió de Shewhart:

As a man, he was gentle, genteel, never ruffled, never off his dignity. He knew disappointment y frustration, through failure of many writers en mathematical statistics to understand his point of view.

He was founding editor of the Wiley Series en Mathematical Statistics, a role that he maintained for twenty years, always championing freedom of speech y confident to publish views at variance with his own.

His honours included:

• Miembro fundador, fellow y presidente de el Institute of Mathematical Statistics;
• Miembro fundador, primer miembro honorario y primer medallista Shewhart Medal de la American Society for Quality;
• Fellow y president de American Statistical Association;
• Fellow de International Statistical Institute;
• Fellow honorario de la Royal Statistical Society;
• Holley medal de la American Society of Mechanical Engineers;
• Doctor Honorario de Ciencia, Indian Statistical Institute, Calcutta.

Quotes

Both pure y applied science have gradually pushed further y further the requirements for accuracy y precision. However, applied science, particularly en the mass production of interchangeable parts, is even more exacting than pure science en certain matters of accuracy y precision.​

Progress en modifying our concept of control has been y will be comparatively slow. en the first place, it requires the application of certain modern physical concepts; y en the second place it requires the application of statistical methods which up to the present time have been for the most part left undisturbed en the journal en which they appeared.​

Shewhart’s propositions​

1. All chance systems of causes are not alike en the sense that they enable us to predict the future en terms of the past.

2. Constant systems of chance causes do exist en nature.

3. Assignable causes of variation may be found y eliminated.

Based upon evidence such as already presented, it appears feasible to set up criteria by which to determine when assignable causes of variation en quality have been eliminated so that the product may then be considered to be controlled within limits. This state of control appears to be, en general, a kind of limit to which we may expect to go economically en finding y removing causes of variability without changing a major portion of the manufacturing process as, for example, would be involved en the substitution of new materials or designs.​

The definition of random en terms of a physical operation is notoriously without effect on the mathematical operations of statistical theory because so far as these mathematical operations are concerned random is purely y simply an undefined term. The formal y abstract mathematical theory has an independent y sometimes lonely existence of its own. But when an undefined mathematical term such as random is given a definite operational meaning in physical terms, it takes on empirical y practical significance. Every mathematical theorem involving this mathematically undefined concept can then be given the following predictive form: If you do so y so, then such y such will happen.​

Every sentence en order to have definite scientific meaning must be practically or at least theoretically verifiable as either true or false upon the basis of experimental measurements either practically or theoretically obtainable by carrying out a definite y previously specified operation en the future. The meaning of such a sentence is the method of its verification.​

In other words, the fact that the criterion we happen to use has a fine ancestry of highbrow statistical theorems does not justify its use. Such justification must come from empirical evidence that it works.​

Presentation of Data depends on the intended actions​

Regla 1. Original data should be presented en a way that will preserve the evidence en the original data for all the predictions assumed to be useful.

Regla 2. Any summary of a distribution of numbers en terms of symmetric functions should not give an objective degree of belief en any one of the inferences or predictions to be made therefrom that would cause human action significantly different from what this action would be if the original distributions had been taken as evidence.

Véase también

• Control chart
• Common cause y special cause
• Analytic y enumerative statistical studies

Publicaciones

Libros

• Shewhart, Walter A[ndrew]. (1917). A study of the accelerated motion of small drops through a viscous medium. Lancaster, PA: Press of the New Era Printing Company. pp. 433 p. OCLC 26000657. LCCN 18007524. LCC QC189 .S5. 

• Shewhart, Walter A[ndrew]. (1931). Economic control of quality of manufactured product. New York: D. Van Nostrand Company. pp. 501 p. ISBN 0-87389-076-0 (edition ??). OCLC 1045408. LCCN 31032090. LCC TS155 .S47. 

• Shewhart, Walter A[ndrew]. (1939). Statistical method from the viewpoint of quality control. (W. Edwards Deming). Washington, The Graduate School, the Department of Agriculture. pp. 155 p. ISBN 0-486-65232-7 (edition ??). OCLC 1249225. LCCN 40004774. LCC HA33 .S45. 

Bibliography

• Deming, W. Edwards (1967) Walter A. Shewhart, 1891-1967, American Statistician, Vol. 21, No. 2. (Apr., 1967), pp. 39-40.
• Bayart, D. (2001) Walter Andrew Shewhart, Statisticians of the Centuries (ed. C. C. Heyde y E. Seneta) pp. 398-401. New York: Springer.
• Fagen, M D (ed.) (1975) A History of Engineering and Science in the Bell System: The Early Years (1875-1925)
• Fagen, M D (ed.) (1978) A History of Engineering and Science in the Bell System: National Service en War y Peace (1925-1975) ISBN 0-932764-00-2
• Wheeler, Donald J. (1999). Understanding Variation: The Key to Managing Chaos - 2nd Edition. SPC Press, Inc. ISBN 0-945320-53-1.

External links

• ASQ Shewhart page
• Walter A Shewhart on the Portraits of Statisticians page.