A stage-wise approach for burden attribution in LCA of multifunctional systems

Abstract

PurposeThe allocation of environmental burdens in multifunctional systems remains one of the most debated methodological challenges in Life Cycle Assessment (LCA). In most applications, allocation is implemented or reported using aggregated allocation rules, which can make burden propagation less explicit and reduce result transparency. This study proposes a generalizable stage-wise bookkeeping and reporting approach that attributes burdens locally at each transformation stage, enabling traceable burden propagation and system-wide reconciliation, while supporting transparent implementation and documentation of allocation choices made under ISO 14044.MethodsThe approach operates on a subdivided process model represented as discrete transformation stages (unit processes) linked by material and energy flows. For each stage, gate-to-gate impacts are calculated and normalized per unit reference flow, and stream burdens are propagated through the network and allocated at multifunctional nodes according to an explicit allocation profile. Single-rationale profiles yield footprints under that rationale, whereas mixed-rationale profiles are reported as allocation-sensitivity scenarios. Accounting closure is checked by reconciling the total system impact with the sum of product burdens. The approach is demonstrated through a sugarcane biorefinery producing sugar, ethanol, and electricity.Results and discussionResults show that the level of system disaggregation, internal-flow representation, and allocation strongly affect the distribution of environmental burdens among co-products. The stage-wise approach enables transparent tracing of burdens through internal flows such as steam and electricity recycling, which are often neglected in conventional LCA models. Compared with global allocation methods, the proposed approach makes the allocation profile explicit and supports traceable burden propagation and system-wide reconciliation (accounting closure) between product-level results and total system impacts. This transparency also facilitates identification of stage-level hotspots and supports more robust sensitivity analyses of allocation factors.ConclusionsThe proposed approach supports the implementation of ISO-aligned operational modelling choices, offering a transparent, algebraic, and scalable approach for multifunctional systems. By integrating burden propagation, allocation, and accounting closure explicitly at the stage level, it enhances the interpretability and reproducibility of LCA studies. Although illustrated through a sugarcane biorefinery, the procedure is broadly transferable and applicable to other complex systems such as energy networks, food processing, and chemical production chains.Recommendations and perspectivesThe procedure is well suited to studies requiring stage-level transparency, especially when a process has shared utilities or recycled streams. The stage-wise representation can be embedded in process-systems analyses (e.g., optimization), and future work should extend it to additional impact categories, uncertainty analysis, and implementation within LCA software environments.