The dairy sector generates substantial greenhouse gas emissions, necessitating specific methods to calculate carbon footprints and identify significant environmental hotspots. Life Cycle Assessment (LCA) is a standard tool for assessing the environmental impacts of dairy supply chains; however, it faces major limitations arising from differences in system boundaries, functional units, allocation techniques, and geographical locations. Hybrid LCA methods address these problems by combining process-based and input–output LCA methods, yet they can be improved to enhance data outputs and system flexibility. Recent advancements in artificial intelligence (AI) and machine learning technologies enable researchers to analyze complex and heterogeneous datasets to generate CO₂-eq emission predictions. Nevertheless, existing AI applications in the dairy sector are largely fragmented and insufficiently integrated within LCA-consistent, system-level frameworks. This study develops a preliminary AI-integrated hybrid LCA system that predicts CO₂-eq emissions across dairy production supply chains. The research findings from this study inform developers in creating an ISO-aligned framework, through a user interface prototype, that enables users to enter specified supply chain and geographical data, view CO₂-eq emissions by location, and identify emission hotspots. The main contribution of this research is the development of a clear system structure that integrates AI prediction methods into established hybrid LCA guidelines. The proposed method provides a foundation for future implementations and supports policymakers and researchers in conducting assessments to enhance dairy-sector sustainability decision-making.

Keywords

Dairy Supply Chain, Hybrid Life Cycle Assessment, Artificial Intelligence, Carbon Footprint, Emission Hotspots.