Coupling big data and life cycle assessment: A review, recommendations, and prospects.

[Display omitted] • Multidimensional links and spatiotemporal variations in LCA remain unresolved. • Multidimensional, dynamic, and spatialized LCA studies are reviewed. • A new idea of fully coupling BD and LCA is proposed, discussed, and prospected. • The BigLCA framework and technical architecture are recommended. • The introduction of BD may render LCA towards scientific decision-making. Life cycle assessment (LCA) is a method that focuses on measuring indicators and making decisions in the environmental dimension. However, its isolation from economic, social, and other dimensions is difficult to identify the interconnections and interactions between multidimensions; its global and static perspectives fail to capture details of spatiotemporal variations effectively. These challenges limit the application of LCA for actual complex systems with multidimensional interweaving and high spatiotemporal heterogeneity. This necessitates an approach that can well quantify multidimensional links and spatiotemporal variations to close the gap. To this end, we reviewed approximately 150 papers recorded in Web of Science and Scopus databases to present the progress in the integration of LCA with different dimensions, and the development of dynamic and spatialized LCAs, as well as identify key challenges. Based on the literature review, we introduced the implications of big data (BD) for LCA to explore a theory for the coupling of BD and LCA. We specifically proposed a universal methodological framework of big life cycle analysis (BigLCA), including four practices: (1) building a spatiotemporal reference system to represent the study object, (2) developing a spatiotemporal inventory analysis scheme based on a modified multi-flow and multi-node model to calculate and integrate massive data, (3) introducing and combining a multi-layer indicator system and system dynamics model to quantify multidimensional indicators and identify their links, and (4) providing spatiotemporal contribution analysis and iterative sensitivity analysis schemes for scientific interpretation. The approach and framework can facilitate the understanding and discussions of the use of BD in LCA, which provides a new approach to improve the accuracy of indicator measurement and the effectiveness and applicability of decision-making. [ABSTRACT FROM AUTHOR]