Livscykelanalys som beslutsstöd i tidiga skeden av byggprojekt - med stöd av artificiell intelligens

Sammanfattning

The construction sector accounts for a significant share of Sweden’s climate impact, and the need for effective tools to analyze and reduce emissions is increasing alongside stricter requirements for climate declarations. Life Cycle Assessment (LCA) is an established method for evaluating environmental impacts throughout a building’s entire life cycle; however, it is currently used primarily to fulfill external requirements rather than as a strategic decision-support tool. The potential to influence climate impact is greatest in the early stages of a project, yet it is precisely in these stages that the use of LCA is most limited. Against this background, the aim of this study is to increase the understanding of LCA as a decision-support tool in the early stages of building projects. The study was conducted using a qualitative research approach based on semi-structured interviews with seven professionals in the construction and real estate sector, representing clients, contractors, and consultants. The interview data were complemented by a literature review of scientific articles, industry reports, and relevant standards, and analyzed using thematic analysis. The results show that the use of LCA in early stages is largely driven by external requirements rather than strategic considerations. The identified barriers can be grouped into three main problem areas: limited availability of information in early stages, a compliance-driven rather than strategic use of LCA, and organizational and technical deficiencies in information flows and system support. An important distinction is that the core issue is not primarily data quality, but data availability data is either missing or not structured in a way that makes it accessible and usable. All respondents identified significant potential for AI to support the LCA process, particularly in automating data collection, structuring data, enabling scenario analysis in early stages, and supporting interpretation and reporting. The analysis indicates that these barriers are structural and embedded in the way construction processes are organized. AI is identified as a tool with clear potential to reduce these challenges, particularly by automating the manual tasks that currently constitute the most time-consuming aspects of LCA. However, the effectiveness of AI is highly dependent on the availability of structured and high-quality data, as well as full transparency in calculation processes. The study highlights a clear gap between the theoretical potential of LCA and its practical application. For LCA to function as an effective decision-support tool, improvements in information flows, increased standardization, and a more strategic organizational approach are required. While AI can contribute to this development, it also introduces risks related to quality assurance and overreliance on automated results. The findings are primarily applicable to the Swedish construction and real estate sector but provide broader insights into how digitalization and AI can support a more sustainable construction process.