Sensitivity and uncertainty analysis of life-cycle assessment based on multivariate regression analysis

Life-cycle Assessment is an iterative procedure where the data to be included should be collected and validated repeatedly to achieve a more accurate picture of environment impacts. Sensitivity analysis and uncertainty analysis are generally recommended to identify key issues for further iterative procedure in a subsequent more detailed LCI. This paper is concerned with the iterative data, using multivariate regression analysis, to find the functional relationship between impact parameters and assessment results. A study to calculate the relative contribution of parameters to the assessment results in regression analysis was carried out. The overall aim of this study was to identify the key sensitive parameters. The paper also discusses the propagation of uncertainties through the regression equation so as to understand how impact parameters influence the assessment results. The level of uncertainty can be derived from the function by means of the partial derivatives. From these analyses, the research offers a set of guidelines to improve data quality. Finally, an example is given to illustrate the methodology.