Towards Autonomous Sustainability Assessment via Multimodal AI Agents

2025-07-22

AI Summary

Traditional life cycle assessment (LCA) of electronic devices is hindered by scarce environmental data across the full “cradle-to-grave” lifecycle, limiting efficiency and accuracy in sustainability evaluation. This paper proposes a multimodal AI agent framework targeting the “cradle-to-gate” stage, integrating text and image understanding, web-based information extraction, and domain-specific data abstraction. It enables direct carbon emission estimation via product description clustering and introduces a novel weighted similarity-driven approach to generate material-specific emission factors, effectively imputing missing material data. Compared to conventional LCA methods requiring weeks, our framework completes end-to-end carbon footprint assessment in under one minute (error ≤19%), with rapid estimation achievable in just 3 ms (MAPE = 12.28%). Emission factor estimation MAPE improves by 120.26%, substantially advancing LCA automation and practical applicability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Interest in sustainability information has surged in recent years. However, the data required for a life cycle assessment (LCA) that maps the materials and processes from product manufacturing to disposal into environmental impacts (EI) are often unavailable. Here we reimagine conventional LCA by introducing multimodal AI agents that emulate interactions between LCA experts and stakeholders like product managers and engineers to calculate the cradle-to-gate (production) carbon emissions of electronic devices. The AI agents iteratively generate a detailed life-cycle inventory leveraging a custom data abstraction and software tools that extract information from online text and images from repair communities and government certifications. This approach reduces weeks or months of expert time to under one minute and closes data availability gaps while yielding carbon footprint estimates within 19% of expert LCAs with zero proprietary data. Additionally, we develop a method to directly estimate EI by comparing an input to a cluster of products with similar descriptions and known carbon footprints. This runs in 3 ms on a laptop with a MAPE of 12.28% on electronic products. Further, we develop a data-driven method to generate emission factors. We use the properties of an unknown material to represent it as a weighted sum of emission factors for similar materials. Compared to human experts picking the closest LCA database entry, this improves MAPE by 120.26%. We analyze the data and compute scaling of this approach and discuss its implications for future LCA workflows.

Problem

• Automating sustainability assessment using multimodal AI agents
• Closing data gaps in life cycle assessment (LCA) for carbon emissions
• Estimating environmental impacts via product similarity and material properties

Innovation

• Multimodal AI agents emulate expert-stakeholder interactions
• Data abstraction extracts info from text and images
• Emission factors generated via weighted material similarity