Abstract

This study develops a comprehensive framework using Data Envelopment Analysis (DEA) to assess the eco-efficiency of AI applications across various sectors. Through an output-oriented DEA model, we evaluate how AI systems balance performance benefits with environmental impacts, incorporating multiple performance indicators and environmental metrics. The research analyzes data from the healthcare, finance, and industrial sectors, using benchmark data and environmental assessments to determine best practices for sustainable AI implementation. The expected results will indicate that the framework effectively identifies eco-efficient AI practices while highlighting limitations in data availability and evolving technological landscapes. The research will contribute to the theoretical understanding of AI eco-efficiency and practical decision-making, offering organizations insights to optimize AI implementations within ESG parameters, ultimately advancing sustainable AI development practices.

Download

Share

COinS