A Data-Driven Evaluation Framework for Quantifying the Impact of Artificial Intelligence on Industrial Process Performance

This study proposes a data-driven evaluation framework to quantify the impact of artificial intelligence (AI) on industrial process performance and enterprise value creation. The framework integrates enterprise value assessment based on the Feltham–Ohlson model with a multi-level performance evaluation framework that incorporates a hybrid Analytic Hierarchy Process (AHP) and Entropy Weight Method (EWM) for indicator weighting, together with Fuzzy Comprehensive Evaluation (FCE) for multi-dimensional aggregation. This integrated approach enables systematic analysis of AI-driven effects from the perspectives of intelligent investment input, operational governance environment, and process output performance. Using panel data from 3515 Chinese A-share listed firms (20,076 firm-year observations) during 2014–2022, a Process Performance Index (PI) is constructed to measure AI-enabled operational capability across resource allocation efficiency, coordination effectiveness, and production performance dimensions. Empirical results indicate that PI is positively associated with abnormal earnings and firm profitability, demonstrating that AI-enabled process capability contributes to sustained enterprise value growth. The findings further show increased digital technology investment intensity, knowledge-based human capital accumulation, and improved data governance conditions, accompanied by enhanced production and service performance. By explicitly integrating AHP–EWM weighting and FCE aggregation within the Feltham–Ohlson valuation structure, the proposed framework provides an interpretable quantitative mechanism linking AI adoption, operational capability development, and enterprise value creation. The results offer practical insights for evaluating intelligent transformation strategies in the context of Industry 5.0 and data-driven industrial development.