AI-BASED CIRCULAR ECONOMY RECOMMENDATION SYSTEM USING DIGITAL TWIN AND EXPLAINABLE ARTIFICIAL INTELLIGENCE

Abstract: The growth in the variety of goods produced in the sectors of electronics, fashion, and household appliances has resulted in environmental problems of resource exhaustion and solid waste management [1]. Regrettably, the current approach to sustainable management of products' life cycle has not managed to establish transparency in consumer choice [2]. In this work, we present, the AI-based Sustainable Product Decision System (AI- SPDS), which combines LDT, Rule-Based Decision Engine, and XAI models to decide on Repair, Reuse, Recycle, and Replace product based on sustainability score, which is a function of Carbon Footprint, Energy Usage, Recycling Capability, and Waste Generation. On 2,340 product datasets in three categories, we achieve an average accuracy rate of 83.1%. Our method is lightweight and easy to understand, making it highly suitable for consumer applications focused on the principles of circular economy (SDG-12).

Keywords: Circular Economy, Digital Twin, Explainable AI, Product Lifecycle, Sustainability Scoring, Rule-Based System, Repair Decision, Waste Reduction