An AI-Based Framework for Assessing Sustainability Conflicts in Medical Device Development
Apala Chakrabarti
TL;DR
The paper tackles the challenge of identifying sustainability conflicts in medical device development by introducing two AI-assisted pipelines that use large language models to classify lifecycle stages and extract conflicts, combined with a structured MCDA to compute a composite sustainability score. It demonstrates the approach through case studies (mercury vs. digital thermometer; eyewear frames and lenses), showing how AI-enabled analysis yields scalable, data-driven decision support for early design. The work highlights how CE graphs and conflict-aware scoring can improve consistency and reduce subjective bias, while acknowledging limitations in automation and generalizability that motivate future research. Overall, the framework offers a principled, data-driven method for balancing environmental, economic, and social objectives in medical device design.
Abstract
Designing sustainable medical devices requires balancing environmental, economic, and social demands, yet trade-offs across these pillars are difficult to identify using manual assessment alone. Current methods depend heavily on expert judgment, lack standardisation, and struggle to integrate diverse lifecycle data, which leads to overlooked conflicts and inconsistent evaluations. This paper introduces an AI-driven framework that automates conflict detection. Machine learning and natural language processing are used to extract trade-offs from design decisions, while Multi-Criteria Decision Analysis (MCDA) quantifies their magnitude through a composite sustainability score. The approach improves consistency, reduces subjective bias, and supports early design decisions. The results demonstrate how AI-assisted analysis provides scalable, data-driven support for sustainability evaluation in medical device development.