Measurement uncertainty evaluation in soil liquid limit testing

Abstract

Abstract Measurement uncertainty provides an indication of the quality of test result and very often also adopted in laboratory proficiency testing activities. Soil mechanical properties testing for example plastic and liquid limits conventionally do not provide uncertainty figure; however due to laboratory accreditation requirement, laboratory is required to provide it in the test report for some specific purposes for example Proficiency testing and quality validation. Conventional approach to derive measurement uncertainty is too difficult or mathematically unable to justify for the figure derived because liquid limit is not obtained experimentally instead it is estimated through linear regression method. Thus, it becomes a problem to report accurate and reliable measurement uncertainty figure in the test report. This research focuses into estimating the water content of liquid limit with relative uncertainty approach through estimation from the regression curve. It suggested the technique for estimation of water content and dry weight resulting in uncertainty component different treatment from conventional approach. Error of regression, a Type A component becomes one of the key components dominating the final answer. This study suggests the control of quality and improvement of measurement uncertainty evaluation through curve fitting observing the value of R2. should not be less than 0.95. This uncertainty figure can be used in proficiency testing and inter-laboratory comparison purpose.