Customer-centric circular economy as-a-service decision-making: Machine learning-driven open innovation in food service

This study addresses a key research gap by developing a machine learning-driven open innovation framework for prioritizing Circular Economy as a Service (CEaaS) measures in the food service sector. Conventional sustainability strategies often fail to integrate dynamic customer preferences with scalable circular economy practices. To bridge this gap, this research employs Random Forest classification and feature selection to assess the impact of sixteen CEaaS measures on customer satisfaction across Indonesian and Hungarian food service markets. The framework is grounded in PRISMA-guided literature review, customer surveys, and machine learning-based decision analytics. Sustainable sourcing, reusable packaging systems, and circular loyalty programs emerge as the most critical CEaaS measures, reflecting global consumer priorities for ethical procurement, waste minimization, and behavior-based engagement. Notably, surplus food redistribution and food waste upcycling services rank consistently high, underscoring growing expectations for visible circularity and social value. The cross-country model validation confirms the framework's robustness and generalizability. By integrating circular economy principles, customer insights, and machine learning, the study advances open innovation theory and provides actionable guidance for sustainable transformation in the food service sector. The findings support Sustainable Development Goals related to responsible consumption, climate action, industry innovation, and global partnerships.