Effective decision-making in plastic reuse and recycling depends on diverse knowledge sources across industry, academia, and policymakers. Nevertheless, fragmented expertise and data silos hinder circular economy initiatives. Moreover, the lack of appropriate tools limits collaboration, creating barriers to innovation. Without collaborative approaches, stakeholders develop isolated models, leading to redundant efforts. Despite the clear benefits of knowledge exchange, a secure and user-oriented framework for sharing insights remains absent. This paper proposes DeCoL-DSS, a conceptual decision support system framework that integrates distributed machine learning models through peer-to-peer collaborative learning strategies. It enables stakeholders to share insights via the model’s outputs or parameters. Functional and non-functional system requirements were derived through a user story approach. The process flow and DeCoL-DSS architecture are demonstrated using an example of incorporating recycled plastics into end products while maintaining quality standards. This work contributes to developing knowledge exchange AI-driven solutions for advancing recycling and waste management and fostering industry-wide collaboration.