Data reconciliation applied to the conformity assessment of fuel products

• Published in: Fuel (Guildford) Vol. 300; p. 120936
• Main Authors: de Oliveira, Elcio Cruz, Lourenço, Felipe Rebello
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 15.09.2021; Elsevier BV
• Subjects: Boiling points; Conformity; Conformity assessment; Consumers; Consumers (storage terminal); Data reconciliation; Diesel and gasoline; Diesel fuels; Flash point; Gasoline; Histograms; Measurement uncertainty; Producers (refinery); Reconciliation; Refineries; Risk of false decisions; Spreadsheets; Sulfur; Sulfur content; Uncertainty; Diesel and gasoline; Conformity assessment; Producers (refinery); Measurement uncertainty; Data reconciliation; Consumers (storage terminal); Risk of false decisions
• ISSN: 0016-2361; 1873-7153
• DOI: 10.1016/j.fuel.2021.120936

[Display omitted] •DR is a good tool to reduce measurement uncertainties.•DR approach improves the reliability of process data.•The use of DR associated to the guard-band concept is an original achievement.•DR is useful to solve discrepancies between producer’s and consumer’s measurements.•A reconciled value leaves no doubt about the conformity assessment of fuel products. The discrepant measurement results regarding the conformity assessment of diesel and gasoline fuels may lead to commercial dispute between producers (refinery) and consumers (storage terminal). The aim of this work was to apply data reconciliation in order to solve the discrepancy between producers’ and consumers’ measurements and allow to make decisions regarding the conformity/non-conformity assessment. Diesel and gasoline samples were analyzed in respect to total sulfur content and flash point and for sulfur content and final boiling point, respectively. Discrepant results between consumers’ and producers’ measurements were subject to data reconciliation. Data reconciliation was perform using a MS Excel spreadsheet. The sheet was divided into five sections: ‘Specification limits’, ‘Producer’s measurement’, ‘Consumer’s measurement’, ‘Reconciled measurement’, and ‘Histograms’. Data reconciliation was useful to solve the discrepancy between producer’s and consumer’s measurements in most of the studied cases, since provided reconciled values inversely weighted as function of measurement uncertainty. In addition, data reconciliation provided a reduced measurement uncertainty.