Naslov
Applying the Bayesian method for evaluating uncertainty in mass calibration

Autori
Bošnjaković, Alen ; Runje, Biserka ; Džemić, Zijad

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Etikum 2021 / - Novi Sad, 2021, 5-8

Skup
International Scientific Conference (ETIKUM 2021)

Mjesto i datum
Novi Sad, Srbija, 02.12.2021. - 04.12.2021

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Uncertainty propagation ; Bayesian method ; prior distribution ; posterior distribution ; MC method ; GUM method

Sažetak
The procedure for evaluating measurement uncertainty by applying the Bayesian method is described in the paper. The Bayesian method differs from the MCS method and GUM method for evaluating the measurement uncertainty. The Bayesian method combines prior knowledge about the measurand with the data collected during the calibration process. The best estimate and coverage interval for the measurand can be calculated from the joint posterior probability distribution obtained from the combination mentioned above. In addition to this, the GUM method, MCS method, and Bayesian method for evaluating the measurement uncertainty are compared for the mass calibration model. It proved that all three methods give the similar results when the Bayesian method is conducted under the noninformative prior distribution.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo

POVEZANOST RADA