Uncertainty Quantification in Life Cycle Assessment for Aviation

Abstract

In the face of climate change and the ongoing environmental crisis, enabling sustainability in aviation through technological solutions has become a priority. To evaluate the potential environmental impacts of such innovations, Life Cycle Assessment (LCA) is a widely adopted method and continues to gain relevance in both research and industry. However, the presence of uncertainties in LCA can significantly influence the trustworthiness of results and limit their usefulness for decision-making. This challenge is particularly relevant when evaluating emerging aircraft concepts, where established and comprehensive data is often scarce. This study addresses these issues by developing a framework for uncertainty and sensitivity analysis tailored to DLR's Lifecyle Analysis Framework for Environment and Economics (LYFE²). LYFE² combines LCA with discrete-event simulation (DES) to model the environmental impacts over an aircraft's full life cycle including production, flight operations, maintenance and end-of-life. Uncertainties in the input parameters are quantified via probability distributions. The semi-quantitative Pedigree approach supports their definition by translating qualitative expert judgement into quantitative standard deviation values. Furthermore, sensitivity analysis and diagnostic diagrams are used to identify the parameters that have the highest influence on the overall quality of the LCA results. To gain insight into how uncertainties in the inputs affect the outputs, Monte Carlo simulations are performed and complemented by a convergence analysis which ensures the reliability of the propagated results. Applying the framework to a case study demonstrates the value of uncertainty awareness when interpreting LCA findings. Presenting results as distributions rather than point values enhances transparency and supports understanding of their informative value. In the context of decision-making, this enables a more differentiated comparison between technologies by examining both the differences in impacts and the associated uncertainty margins. Moreover, juxtaposing the sensitivity and uncertainty of individual input parameters helps to prioritise subsequent data improvement efforts, i.e., attention can be directed to those areas where a reduction in input uncertainty would yield the greatest benefit to the reliability of the assessment. The developed framework establishes a structured approach to uncertainty quantification in LCA, in accordance with ISO 14040 and 14044 standards. In a broader context, this work contributes to improving the reliability and transparency of LCA in aviation, therefore supporting evidence-based decision-making in the development of sustainable technologies.