Uncertainty analysis in life cycle inventory. Application to the production of electricity with French coal power plants

Uncertainty assessment in LCAs is an important aspect for decision-makers to judge the significance of differences in product or process options. Stochastic models (e.g., Monte Carlo) are tools used in Life Cycle Inventory (LCI) to compute the uncertainty of cumulative emissions and resource requirements. However, one main problem when applying such models is the large number of unit processes that make up a product system. In this paper a combination of qualitative and quantitative approaches in uncertainty assessment is proposed for an efficient assessment of uncertainty. The qualitative assessment of data quality relies on data quality indicators, whereas the quantitative assessment uses Monte Carlo simulation. The effort to select accurate probability functions concentrates on data with a significant contribution to the cumulative results1 and/or with a high uncertainty. The probability function of selected data is estimated using different techniques, depending on the amount of information available. Some results are presented by applying the method on selected French coal-based electricity.