Guillaume Majeau-Bettez
CIRAIG, Polytechnique Montreal; and NTNU
Guillaume is a postdoctoral researcher, jointly employed by the CIRAIG, Polytechnique Montreal, and the IndEcol Programme at the Norwegian University of Science and Technology (NTNU). His research is focused on the coherent integration of industrial ecology tools and methods to yield a more comprehensive and accurate guidance toward a more sustainable socioeconomic metabolism.
Process-based lifecycle assessments (LCAs) rely on physical data that are as detailed and specific as possible. However, this level of detail means that only a fraction of the total production system can feasibly be described, and incomplete and inconsistent system boundary definitions therefore constitute a key weaknesses of process-based LCAs. This truncation adds uncertainty to the comparative assessment of products with differing value chains. It also introduces a negative bias in absolute environmental impact estimates and complicates the comparison between independent studies.
Multiple hybridization approaches have been proposed to complement the detail of process-based LCA descriptions with the completeness of the (more aggregated) economic input-output (IO) tables, thereby partly resolving the issue of truncation. However, these hybridization techniques are highly data intensive and complicated by the need to work in both monetary and physical units. Hybrid LCA-EEIO analyses remain rare and restricted to individual case studies, rather than reaching mainstream practice or becoming the norm in database development.
There is evidence that complete sectors of the economy are underrepresented in LCA databases, notably the ecoinvent database. Consequently, at present, even if a hybrid LCA-EEIO analysis resolves truncation issues in the foreground, the reliance on LCA databases for background processes still introduces truncation and inconsistent system boundaries.
The current research project aims to resolve these issues: we aim to develop a streamlined hybridization process that is lean on data and is facilitated by open-source software. This approach is then employed to hybridize not only case studies but also the complete ecoinvent database, combining it with the multiregional IO database, Exiobase. To this end, we address multiple practical and methodological issues, notably pertaining to correction for double-counting and price heterogeneity.
By combining elements from the path-exchange and the price-model approaches to tiered-hybrid analysis, and by introducing numerous simplifying assumptions, we propose a streamlined workflow for a credible hybridization with minimal data requirements.
When LCA processes respect mass and elemental balances, the correction for double counting is greatly simplified. In such cases, any input from an IO sector that is deemed likely to contribute to the mass of the product can be assumed to be already covered in the LCA part of the analysis. These are then nullified, which leaves inputs from different categories of IO sectors that come as a complement to process LCA descriptions: services, waste treatment, infrastructure, etc.
The joint manipulation of LCA and IO databases, the categorization of IO sectors, the correction for double-counting, and the disaggregation of sectors with heterogenous prices are all facilitated and partly automated by an open-source, documented, flexible and unit-tested software, programmed in Python: pyLCAIO.
Ecoinvent and Exiobase are proving to be highly compatible and complementary databases: both represent value chains spanning international borders in a multiregional manner, both have a strong focus on the energy sectors and on resource use. Their ongoing hybridization should yield an optimally detailed and complete description of the technosphere, along with global insights into the issue of truncation in LCA practice in different sectors of interest.
• Environmentally and socially-extended input-output analysis , • Life cycle sustainability assessment , • Sustainable consumption and production
