All measured data and models contain errors, and Life Cycle Inventory (LCI) data are no exception, particularly since LCI data are usually collected and combined from diverse sources, which may be unreliable, incomplete or unverifiable. Since the underlying or noise free LCI data must satisfy the laws of conservation, their quality may be improved by imposing these laws on measured data. Therefore, data rectification must be applied to inventory data for enhancing the accuracy and reliability of LCA. In this paper, the procedure of data acquisition and rectification is suggested with a reliability measure to evaluate the usefulness of additional data. The reliability measure can be used for deciding the focus of future data collection and the accuracy needed in additional data to justify the extra effort or cost.

We suggest practical approach for data acquisition decision and data rectification. If we meet an opportunity to get additional information, we should calculate the cost and benefit of the new information. The methodology starts by determining if we purchase the additional data. For this decision, we propose a reliability measure for the data quality. The proposed estimator can be used for quantifying the relative improvement in the LCA results due to the use of such additional information. Since this evaluation can be done without having access to data, it can also be used to determine whether it would be worth spending the time or money for obtaining data of a given accuracy. Inversely, this approach could also be used for determining the type of additional information that should be obtained to achieve a desired overall accuracy of LCA.