Back to AI Research

AI Research

LCAi: Life Cycle Assessment with big data fusion an... | AI Research

Key Takeaways

  • LCAi: Life Cycle Assessment with big data fusion and retrieval-augmented generation-assisted interpretation Life Cycle Assessment (LCA) is a standard method...
  • To operationalise large language models in LCA interpretation, a perspective fusion RAG architecture was developed, covering academic, industry, public discourse, and European union (EU) funding datasets.
  • The framework is demonstrated through a hydrogen-enabled diesel reduction use case in an Italian apple production facility using GPT-5 nano as the reasoning model.
  • Overall, the structured retrieval and constrained synthesis are designed to mitigate the risk of hallucination while preserving cross-domain diversity.
  • This proof-of-concept demonstrates how AI-assisted, evidence-grounded interpretation can support implementation-oriented decision-making beyond conventional LCA studies.
Paper AbstractExpand

The interpretation phase of life cycle assessment often lacks structured mechanisms for translating quantified improvement opportunities addressing environmental hotspots into actionable strategic pathways under technological, social, and policy uncertainty. To overcome this limitation, this study introduces a perspective-conditioned retrieval-augmented generation framework for LCA interpretation, where a multi-perspective retrieval and controlled synthesis is incorporated in the artificial intelligence (AI)-assisted LCA. To operationalise large language models in LCA interpretation, a perspective fusion RAG architecture was developed, covering academic, industry, public discourse, and European union (EU) funding datasets. Our approach comprises three steps: (1) a scenario anchor defining system boundaries and decarbonization targets, (2) a set of perspective-specific micro-queries with constrained retrieval, and (3) a neutral synthesis step integrating only ledger-stored outputs without further retrieval. The framework is demonstrated through a hydrogen-enabled diesel reduction use case in an Italian apple production facility using GPT-5 nano as the reasoning model. Overall, the structured retrieval and constrained synthesis are designed to mitigate the risk of hallucination while preserving cross-domain diversity. The approach presented can support more disciplined translation of impact results into strategic pathways and opens up new avenues for the use of advanced AI tools in LCA studies, particularly those focused on technologies that could be deployed at scale. This proof-of-concept demonstrates how AI-assisted, evidence-grounded interpretation can support implementation-oriented decision-making beyond conventional LCA studies.

LCAi: Life Cycle Assessment with big data fusion and retrieval-augmented generation-assisted interpretation
Life Cycle Assessment (LCA) is a standard method for measuring the environmental impact of products, but it often struggles to turn technical data into clear, actionable strategies for real-world decision-making. This paper introduces a new framework called LCAi, which uses Artificial Intelligence to bridge the gap between complex environmental data and practical implementation. By combining traditional LCA results with diverse, real-world information—such as academic research, industry trends, public opinion, and government policy—the framework helps stakeholders create evidence-based roadmaps for sustainability.

A Multi-Perspective Approach

The core innovation of this framework is its use of "perspective-conditioned" retrieval. Instead of relying on a single source of information, the system pulls data from four distinct domains: academic literature, industry profiles, public discourse (such as social media and forums), and EU funding policies. By separating these sources, the system can provide a well-rounded view of a project. For example, while academic data might highlight technical efficiency trade-offs, industry data focuses on market readiness, and public discourse reveals potential social concerns or safety perceptions.

How the Workflow Works

The framework operates through a structured, three-step process designed to ensure accuracy and prevent the AI from "hallucinating" or making up facts:

  1. Scenario Anchoring: The process begins by defining a specific "anchor"—a set of boundaries, goals, and environmental hotspots derived from a standard LCA. This ensures that all subsequent AI analysis remains focused on the specific project at hand. 2. Perspective-Specific Retrieval: The system uses a Retrieval-Augmented Generation (RAG) architecture. It searches through the four data domains to find relevant information based on the project’s needs. It uses strict constraints to ensure the AI only uses the retrieved evidence to answer questions. 3. Neutral Synthesis: Finally, the system takes the outputs from all perspectives and combines them into a single, consolidated roadmap. Crucially, this final step does not perform new searches, ensuring the final report is based only on the audited, ledger-stored outputs from the previous steps.

Practical Application

To test this approach, the researchers applied it to a case study involving the use of hydrogen to reduce diesel consumption in an Italian apple production facility. Using a model called GPT-5 nano, the framework successfully generated an auditable 2030 roadmap. The results demonstrated that the AI could synthesize complex, cross-domain information into a coherent plan, identifying funding opportunities from EU policy while simultaneously addressing potential safety concerns raised in public discourse.

Considerations for Future Use

This proof-of-concept shows that AI can act as a powerful assistant for interpreting LCA results, moving beyond simple data analysis toward strategic implementation. However, the authors note that this is an initial step. Future research will focus on expanding the types of data integrated into the system—such as sustainability reports and official legislation—and adapting the framework to handle more complex, multi-objective scenarios. The goal is to create a more robust, standardized tool that can help organizations navigate the uncertainties of technology, policy, and social acceptance when deploying sustainable solutions at scale.

Comments (0)

No comments yet

Be the first to share your thoughts!