03 November 2015: Stephan Conrad

Who is not familiar with the famous statement about the art of measuring: “The one who measures measures crap“. The most crucial question for any type of measurement is about accuracy and precision. The frst procedure ensuring measurement accuracy was developed in ancient Egypt. As early as 2500 B.C. they carved a “royal cubit master“ out of a block of granite to measure the water level of the Nile and to gain comparable results. On a visit to Paris, you should also keep your eyes peeled for the marble metre bar. You may fnd two of these 18th century material measures on house walls. The Parisian population, especially merchants, compared the length of their measuring sticks to these marble metres. This was also the time when the first provisional metre bar was constructed. Only a few years later, the frst mètre des
archives platinum bar standard was also produced there; it was fnally replaced by a platinum-iridium bar another 100 years later. In 1960, the prototype metre was defned by the wavelength of light from a specifed
transition in krypton-86 in vacuum and in 1983; the metre became the length of the path travelled by light in vacuum during a specifed time interval. All these changes did not always intend to provide a new defnition of the size of a metre but tried to increase its precision.

GUM - Guide to the expression of uncertainty in measurement

It is even more surprising that the SI system has not had any uniform rules as to measurement “accuracy“ yet at the time when the length of the metre was defned – a length that we still apply today. Even though the international Committee for Weights and Measures (CIPM) had already requested a recommendation on the calculation of measurement uncertainty from the international Bureau of Weights and Measures (BIPM), ISO was not entrusted with preparing a detailed guideline until 1986. The “Guide to the expression of uncertainty in measurement“ (GUM) based on the BIPM recommendation was published in 1995.

The beginnings of measurement system analysis

At that time, the automotive industry also tried to introduce SPC in production, including the operator self-inspection as is common practice today. SPC measuring stations close to the machines were provided with various measuring instruments in order that operators are able to take required measurements directly. However, the automotive industry hardly considered the question at all whether the respective measuring instruments were able to fulfl the corresponding measuring task with suffcient accuracy...