Brightcon 2026, hackathon and courses

Industrial Ecology increasingly aims at connecting complementary assessment frameworks rather than limiting analysis to conventional Life Cycle Assessment (LCA). In this context, we propose a framework to assess what we define as intrinsic circularity, i.e., the direct circulation of physical flows through and across product systems (1).

The framework questions the conventional definition of product system boundaries in LCA. Product systems frequently exclude so-called cut-off product flows, either due to cut-off criteria application (2) or multifunctionality resolution strategies (3). We argue that these flows may nevertheless represent important carriers of physical content entering or leaving the studied system, and therefore should be accounted for when evaluating circularity.

To address this issue, we developed a workflow based on the Python library Brightway. Two Python modules were implemented to identify and extract cut-off product flows from LCA databases. A second module (circularity_calculator.py) derives circularity indicators directly from the Life Cycle Inventory (LCI). The approach requires translating heterogeneous physical units (e.g., kg, MJ, kWh, m3) into common equivalent metrics such as kg-eq or MJ-eq. To achieve this, the framework uses recorded flow properties available in databases such as ecoinvent v3 (e.g., densities or calorific values).

For the presentation, the proposed framework is illustrated through a simple Lithium-ion battery storage system example. A multifunctional battery use process generates both an energy storage service and an end-of-life battery flow. Under traditional cut-off modelling, the end-of-life flow remains outside the circularity accounting framework. We extend the intervention matrix to include cut-off product flows as resource flows contributing to circularity measurements. The resulting framework allows the derivation of circularity and inefficiency indicators. In the example case study, the proposed method captures circular flows ignored by the traditional LCI formulation, including the circulation of stored energy and end-of-life batteries. The approach also highlights the importance of assigning physical properties to flows: without conversion factors such as water density or fuel calorific value, physically meaningful aggregation becomes impossible.

The framework was further applied to 1000 randomly selected product systems from the ecoinvent v3.12 cut-off system model. Figure 1 presents inefficiency indicators expressed in both mass-equivalent and energy-equivalent units. Preliminary observations suggest that inefficiency values expressed in MJ-eq remain systematically low, likely because heat dissipation and thermal losses are insufficiently recorded in current LCA databases. This highlights both the potential and the present limitations of database structures for intrinsic circularity assessment.

More broadly, this work opens a discussion on the allocation choice to maintain physical balance, and of the utilitarian nature of LCA and questions the centrality of the functional unit when analysing the circulation of physical content across interconnected product systems.

Figure 1 (4): Inefficiency indicators derived from 1000 randomly selected product systems from the ecoinvent v3.12 cut-off system model, expressed in kg-eq and MJ-eq. Results suggest that inefficiency values expressed in MJ-eq remain systematically low, likely due to insufficient recording of heat dissipation and thermal losses in current LCA databases.

Post scriptum
The attached abstract includes the simple numerical example detailing the proposed methodology. The corresponding calculations and implementation are available in the accompanying GitHub repository.

References:
(1) Michael Saidani et al. “A taxonomy of circular economy indicators”. In: Journal of Cleaner Pro-duction 207 (2019), pp. 542–559. ISSN: 09596526. DOI: 10.1016/J.JCLEPRO.2018.10.014. arXiv: 1901.02709.

(2) ISO. Environmental management — Life cycle assessment — Principles and framework. ICS : 13.020.10 13.020.60. Geneva, Switzerland, July 2006. URL: https://www.iso.org /standard/37456.html.

(3) Reinout Heijungs and Sangwon Suh. The Computational Structure of Life Cycle Assessment. Vol. 11. ISBN: 978-90-481-6041-9. Dordrecht: Springer Netherlands, 2002. DOI: 10.1007/978- 94-015-9900-9. URL: http://link.springer.com/10.1007/978-94-015-9900-9
(visited on 09/04/2024).

(4) Figure 1: https://github.com/LouisFreboeuf/circularity_lca/blob/main/results/plots/CI-kg_CI-mj/pc_dup_bw-25_ecoinvent-3.11-cutoff_fm-fe_1000uprs/eta-_hybrid_scale_by_section.svg