A survey of Life Cycle Inventory database implementations and architectures, and recommendations for new database initiatives

Purpose Barriers to interoperability pose a serious challenge in the domain of LCI database development and prevent effective data sharing between LCA practitioners. Through analysis of existing LCI database resources and technology, a set of recommendations was developed to promote interoperability, and thus utility, in developing new LCI data resources, including the forthcoming Canadian Agri-food Life Cycle Data Centre. Methods A review of published literature, user documentation, gray-literature technical reports, and conference proceedings, as well as interviews with LCI database experts, was used to determine current common/best practices in the LCI database field. On this basis, core interoperability criteria and barriers were identified, and potential solutions were explored. A set of minimum, generalizable recommendations for LCI database development was then identified, aimed at ensuring that these core interoperability criteria are met. These recommendations were further informed by current popular practice in computer science and database-driven web application fields. Results and discussion Data exchange format and nomenclature were identified as core interoperability criteria. Both the ILCD and EcoSpold formats and nomenclatures were found to be in widespread use, and are recommended for support to ensure interoperability with the most widely used LCI databases. It was also found that technical implementations of LCI databases largely followed practice in the database-driven web application field, and likely has little effect on interoperability between databases. In addition, third-party data providers and networks such as OpenLCA Nexus, ecoinvent, and the UNEP GLAD network were identified as potential solutions to interoperability challenges, allowing for greater distribution and interoperability of data, with minimal expense for early adoption. Conclusion A variety of potential solutions were found for interoperability concerns in LCI database development. A final set of five recommendations for format, nomenclature, third-party providers, third-party initiatives, and technical implementation was developed, which can help to ensure a minimum level of interoperability in the development of new LCI database resources. These recommendations will be implemented in the development of the Canadian Agri-Food Life Cycle Data Centre.