Making Climate Data Actionable in Energy Investment Projects

Abstract

Göteborg Energi has established ambitious targets to reduce the climate footprint of its procurement by 90% by 2030. While Life Cycle Assessment (LCA) is an established methodology for quantifying environmental impact, its practical integration into investment decisions is often constrained by organizational and data quality barriers rather than technical limitations. This thesis investigates how Göteborg Energi’s working approach for climate data can be developed to support both early investment decisions, LCA-based follow-up and declarations in larger investment projects. Using a mixed-methods approach structured around the DMAIC framework, a retrospective pilot LCA was conducted on the biomass-fired combined heat and power plant Rya BKV, delimited to the main supplier Valmet’s scope of delivery. This were combined with semi-structured interviews, questionnaires, process observations, and supplier dialogue. The analysis reveals three categories of barriers. Process-related barriers include late and unclear requirements specification in procurement. Data quality barriers including a strong reliance on generic emission factors due to limited availability of product-specific Environmental Product Declarations (EPDs). Organizational barriers arise from unclear allocation of responsibilities between project management, procurement, and the environmental function. In response, an improved working method is proposed that separates climate data into two distinct flows: a limited decision-support flow for use in tender evaluation and a comprehensive follow-up flow post-award, utilizing a standardized supplier data template. The method is reinforced by a shared terminology structure for climate data types, explicit allocation of responsibilities, and a four-level fallback process for missing emission factors. The result is a scalable and structured approach that bridges the gap between early climate screening and rigorous LCAbased follow-up, without assuming complete product-specific data availability at all project stages.