The paper is a review on the worst case uncertainty (WCU) concept, neglected in the Guide to the Expression of Uncertainty in Measurements (GUM), but necessary for a correct uncertainty assessment in a number of practical cases involving distribution with compact support. First, it is highlighted that the knowledge of the WCU is necessary to choose a sensible coverage factor, associated to a sensible coverage probability: the Maximum Acceptable Coverage Factor (MACF) is introduced as a convenient index to guide this choice. Second, propagation rules for the worst-case uncertainty are provided in matrix and scalar form. It is highlighted that when WCU propagation cannot be computed, the Monte Carlo approach is the only way to obtain a correct expanded uncertainty assessment, in contrast to what can be inferred from the GUM. Third, examples of applications of the formulae to ordinary instruments and measurements are given. Also an example taken from the GUM is discussed, underlining some inconsistencies in it.

On the worst case uncertainty and its evaluation / Fabbiano, Laura; Giaquinto, Nicola; Savino, Mario; Vacca, Gaetano. - In: JOURNAL OF INSTRUMENTATION. - ISSN 1748-0221. - 11:02(2016). [10.1088/1748-0221/11/02/P02001]

On the worst case uncertainty and its evaluation

FABBIANO, LAURA;GIAQUINTO, Nicola;SAVINO, Mario;VACCA, Gaetano
2016-01-01

Abstract

The paper is a review on the worst case uncertainty (WCU) concept, neglected in the Guide to the Expression of Uncertainty in Measurements (GUM), but necessary for a correct uncertainty assessment in a number of practical cases involving distribution with compact support. First, it is highlighted that the knowledge of the WCU is necessary to choose a sensible coverage factor, associated to a sensible coverage probability: the Maximum Acceptable Coverage Factor (MACF) is introduced as a convenient index to guide this choice. Second, propagation rules for the worst-case uncertainty are provided in matrix and scalar form. It is highlighted that when WCU propagation cannot be computed, the Monte Carlo approach is the only way to obtain a correct expanded uncertainty assessment, in contrast to what can be inferred from the GUM. Third, examples of applications of the formulae to ordinary instruments and measurements are given. Also an example taken from the GUM is discussed, underlining some inconsistencies in it.
2016
On the worst case uncertainty and its evaluation / Fabbiano, Laura; Giaquinto, Nicola; Savino, Mario; Vacca, Gaetano. - In: JOURNAL OF INSTRUMENTATION. - ISSN 1748-0221. - 11:02(2016). [10.1088/1748-0221/11/02/P02001]
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11589/1623
Citazioni
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact