Asymmetric Systematic Uncertainties in the Determination of Experimental Uncertainty

摘要

A new method is presented for determining a 95/100 confidence uncertainty interval for an experimental result when some of the measured variables have asymmetric systematic uncertainties. The technique is compared with the approximate method given in the American National Standards Institute/American Society of Mechanical Engineers (ANSI/ASME) standard on measurement uncertainty. Monte Carlo simulations are used to show that the new method provides consistent coverages of about 95/100 whereas the ANSI/ASME method provides inconsistent confidence intervals.