A1256
Title: Triple-win performance measurement for sustainable supply chains: A Markov-switching decision trees approach
Authors: Fabio Demaria - University of Modena and Reggio Emilia (Italy) [presenting]
Maddalena Cavicchioli - University of Modena and Reggio Emilia (Italy)
Ulpiana Kocollari - University of Modena and Reggio Emilia (Italy)
Federico Bertacchini - University of Modena and Reggio Emilia (Italy)
Abstract: The assessment of sustainability within supply chains has become essential for performance measurement, particularly in identifying the primary objectives that sustainability metrics should address. While existing literature often highlights the importance of transparency regarding social and environmental impacts, the core goals of sustainability management are to objectively assess and mitigate risks while improving performance. A significant challenge in sustainability performance measurement and management research is addressed: identifying trade-offs and achieving triple-win outcomes by analyzing economic, social, and environmental data across the supply chain. Advanced techniques that integrate parametric and non-parametric machine learning tools are employed, specifically Markov-switching decision trees, which combine decision tree methods with time-series modeling. This approach is well-suited for uncovering latent patterns and diverse sustainability performance combinations based on various stakeholder expectations. In fact, the states are capable of identifying evolving patterns in sustainability performance that successfully respond to diverse stakeholder priorities. Findings effectively measure sustainability issues through the triple bottom line pillars and introduce a new strategy for sustainability reporting.
