Prescriptive Decision-Making for Sustainable Production Management : An Overall Sustainable Equipment Effectiveness (OSEE) Framework Using Causal AI

Abstract

In response to the growing challenges posed by climate change and demographic shifts, industrial operations must move beyond traditional productivity metrics such as Overall Equipment Effectiveness (OEE). While OEE is a valuable key performance indicator, it fails to account for the ecological, social, and economic dimensions essential for long-term sustainability. This paper introduces an Overall Sustainable Equipment Effectiveness (OSEE) framework, designed to integrate sustainability factors into operational performance measurement, enabling a holistic assessment and optimization approach. Key sustainability factors and their interrelationships are identified through an extensive literature review and subsequently validated by industry experts to ensure practical relevance and applicability to real-world operational settings. To address the complexity of these interconnected factors, causal AI methods, in particular Dynamic Bayesian Networks (DBN) are employed. DBN allow a qualitative understanding of sustainability interrelationships (cause-effects) and enable a quantitative optimization of sustainability impacts on operational efficiency. The proposed OSEE framework offers a structured approach for balancing productivity with environmental and social factors, equipping decision-makers with insights for informed sustainable operational strategies. This research contributes to the broader agenda of twin transformation, aligning digitalization and sustainability, and provides a foundation for building resilient, future-ready industrial operations.