Integrating machine learning into life cycle assessment: Review and future outlook

Life Cycle Assessment (LCA) is widely used to quantify environmental impacts but often faces data gaps, heterogeneous practices, and limited timeliness. This review examines how machine learning (ML) can strengthen LCA across all four phases—goal & scope, life cycle inventory (LCI), life cycle impact assessment (LCIA), and interpretation—while providing a reproducible bibliometric map of recent research. We performed a bibliometric search and keyword co-occurrence visualization (VOSviewer) and organized the literature by LCA phases. We highlight actionable opportunities: NLP-assisted scope definition, probabilistic imputation and uncertainty quantification for LCI, surrogate and hybrid models for LCIA, and calibrated, decision-oriented interpretation. Compared with prior reviews, we (i) deliver phase-specific guidance instead of generic lists, (ii) extend coverage to recent work with reproducible bibliometrics, and (iii) foreground early-phase opportunities that remain under-explored. These insights—together with open materials for reuse—aim to make LCA more data-robust, transparent, and actionable in research and practice.