Method to decompose uncertainties in LCA results into contributing factors

Understanding uncertainty is essential in using life cycle assessment (LCA) to support decisions. Monte Carlo simulation (MCS) widely used characterize the variability LCA results, be them inventory (LCI), category indicator normalized or weighted results. In this study, we present a new method decompose MCS results into underlying contributors logarithmic mean Divisia index (LMDI) decomposition with case study on natural gas focusing two impact categories: global warming and USETox human health impacts. First, after each run of MCS, difference simulated deterministic decomposed LMDI method, which returns contribution factor run. After repeating for 1000 runs, statistical properties contributions by are analyzed. The quantifies variables, such as characterization factors LCI items, overall result, characterized presented can variabilities LCI, characterized, normalized, factors, normalization references, weighting any subset them. As an illustrative example, was conducted, were items factors. show that phases contribute 65% 35%, respectively, result warming. For category, LCIs 32% 68%, uncertainty. particular, methane emissions contributed most uncertainties impact, while chromium identified main driver gas. Using approach, practitioners under setting, help prioritize parameters need further refinement reduce reliably estimates variables large without computational resources, it applied stage calculation including weighting, other fields than material flow analysis risk assessment.