"Uncertainty" is a newly developed scale to convey the reliability of measurement data and has been used since 1990s. In the past, the concepts of "error" or "accuracy" were used to express the reliability of measurement. However, the practical application of these concepts widely varied, depending on the fields of technology and countries. Thus, the International Committee of Weights and Measures (CIPM) took the initiative in developing an integrated method for evaluating and describing the reliability of measurement data. The collaborated efforts resulted in a 1993 guide book co-published by the seven main international organizations of metrology --- "the Guide to the Expression of Uncertainty in Measurement." This guide book is often called "GUM" taking initial letters from the original title.
     GUM defines "uncertainty" as a mean to express the degree of ambiguity in our knowledge gained by measurement and explains the detailed procedures of quantitative evaluation. An uncertainty value is based on two different ways of evaluation. The Type A evaluation uses the conventional statistical analysis of the standard deviation. The Type B evaluation uses the estimation of a value comparable to the standard deviation from a variety of information other than measurement data. The results from those two types of evaluation are then put together to get a combined uncertainty.
     "Uncertainty" is now used in various technical and academic documents in which the reliability of measurement data play a significant role. The evaluation of uncertainty is an imperative requirement for standard specifications such as ISO 9000 (for quality systems) and ISO 17025 (for general requirements on the capability of calibration and testing laboratories).