Assessment of environmental product declaration and databases: Towards ensuring data quality assurance practices
Oludolapo Ibrahim Olanrewaju, Wallace Imoudu Enegbuma, Michael Donn, Oluwatobi Nurudeen Oyefusi
Abstract
Environmental product declaration (EPD) is a vital tool for sustainable decision-making. However, limited studies have considered electronic databases' importance in shaping EPD adoption's future. The rise in EPDs globally has raised concerns about data quality assurance issues. Therefore, assessing the current EPD and its databases is important to provide a robust pathway for ensuring data quality assurance practices. This study adopted a mixed research approach to address the concerns, including desk research, focus group discussion, and the use of natural language processing (NLP) for analysis of sample EPDs. The desk research includes data scrapping of EPD data from the ECO Platform and expert ranking agreement based on pre-defined attributes. The study revealed a need to improve the current EPD databases to enhance user experience and the structure of construction EPDs due to the poor association between EPDs published by different programme operators. It is also important to note that the top three EPD databases based on expert ranking agreement are: ECO Platform, EPD Norway, and EPD Ireland. The study proposed a conceptual framework and system architecture aimed at improving data quality assurance practices. The findings from this study have direct implications for EPD programme operators, manufacturers, EPD data users, and policy-makers. • The current EPD databases have evidence of weaknesses and strengths. • The review and data scrapping provided a clear structure of the ECO Platform connection with other standalone databases. • There is a significant difference in terms of contents structure between the EPDs published by different programme operators. • The incorporation of performance and data quality assurance metrics into EPDs could enhance their usability. • A best practice framework and system architecture to improve EPD databases are proposed.