G-41 (Buildings) / G-39 (Products) Reproducibility

*Aspect*		*G-41 (Buildings) / G-39 (Products) Reproducibility*
*Description*		Reproducibility is an important aspect of the reporting step in life cycle assessment. According to ISO 14044 and the ILCD Handbook, it is characterized by providing comprehensive documentation on data, assumptions, calculation rules etc. that would allow an independent reviewer to adequately reproduce the results and the corresponding interpretation of any LCI and LCA study. Because of different data sources and confidential datasets, reproducibility may sometimes be difficult to achieve. In addition, the use of LCA in the building sector leads to the development of LCA software that may use different data, assumptions and calculation rules for a national context. How can reproducibility of LCA studies in the building sector be ensured without excessively increasing the time needed to complete a study?
*related study objective*		☒ stand-alone LCA		☒ comparative assertion
*related study phase*		☐	☐	☐	☐	☒
		goal and scope definition	inventory analysis (LCI)	impact assessment (LCIA)	interpretation	reporting
*relevant for*
	new buildings	existing buildings	building products	screening LCA	simplified LCA	complete LCA
*Provisions*	The description of all aspects related to the LCA study should be as transparent as possible.For confidential data, the assumptions should be made available to the independent critical reviewer (if relevant for the review process).For building LCA studies, the documentation may not be required at the same level of detail as for other LCA studies, owing to the use of predefined data or models (e.g. aggregated generic or EPD datasets with only impact indicators).
*Rules from:*	*EN 15978: 13 Verification of results EN 15804:* 9 Verification and validity of an EPD *ILCD: Provisions: 6.2.2 Reproducibility* I) SHALL – Documentation for reproducibility II) MAY – Accompanying documentation process III) SHALL – Confidential information Provisions: 10.2 Reporting principles I) SHALL – Reproducibility and target audience to guide reporting
*Guidance*	1) General guidance for ensuring reproducibility of product and building LCA studies The ILCD Handbook provides detailed guidance on reporting. To ensure the best reproducibility of results, detailed reporting should be provided. This is especially true for aspirations concerning the quality of complete LCA studies. So, for instance, the author(s) of a complete LCA study should review whether the documentation using the EeBGuide reporting templates provided is sufficient to allow third parties to reproduce their study. If not, the practitioner should extend the documentation by considering the ILCD LCA report template and LCI reference dataset format. 2) Specific guidance for confidential product or building LCA studies It is not always possible to make all the assumptions available to the public, e.g. for a comparative LCA study of two structural systems, owing to confidential industry data. According to the ILCD Handbook, there is a need to balance reproducibility and confidentiality. In general, an independent external reviewer (e.g. for an LCA study) or verifier (e.g. for EPD) is a sound option to ensure that the results of comparative LCA studies (disclosed to the public) are reproducible. In this special case, the confidential data are made available only to the reviewer for e.g. raw data or unit process data in the case of a product LCA. 3) Specific guidance for building LCA studies The LCA methodology cannot be applied with the same level of detail for building LCA studies. Various simplifications are applied, e.g. by using only LCIA indicators instead of full LCI or unit processes. In this context, reproducibility remains as important as for product LCA. It is very important in a national context, when e.g. several LCA software tools exist on the market. Possible ways of achieving reproducibility for building LCA studies include: – harmonization of physical building description data (see the corresponding aspect); – choice of a common national set of generic LCA data (that can be used for early design stages). In addition, some specific requirements are needed for each LCA software: – Provide dedicated user manuals. – Document, in a transparent way, the assumptions of the LCA software (data, calculation rules, expression of results). – Facilitate data selection with e.g. a predefined building description. Reproducibility can be assessed in various ways. For example, training tests can assess the reproducibility among different LCA practitioners using the same LCA software and case study. Here, the influence of the practitioner is being assessed. Conversely, the same LCA practitioner can model the same case study using two different building LCA software tools. Here, the databases and methodological assumptions of the software are being assessed. One important aspect of reproducibility is the use of common datasets. Unfortunately, in some national contexts, different databases (e.g. generic and EPD data) can be used, answering different goals. In this specific case, data analysis tools may need to be developed to assess the differences in LCA data on building products and processes, in addition to the reproducibility tests explained above. Some guidelines on this topic, and on implementation in the French context, can be found in [Lasvaux 2011].