FRACTIONAL-ORDER MATHEMATICAL MODELLING AND AI-DRIVEN OPTIMIZATION OF E-WASTE RECYCLING AND ASSET RECOVERY

Abstract

The accelerating growth of electronic waste has created urgent pressure on recycling systems to recover valuable materials efficiently while minimizing environmental harm. Conventional optimization approaches often overlook the fact that ewaste processing is not memoryless: past operational decisions, delays, and inefficiencies continue to influence current system performance. In this work, we propose a hybrid framework that combines fractionalorder mathematical modelling with artificial intelligence (AI) to optimize ewaste recycling and asset recovery. Fractional calculus is used to describe longrange temporal dependencies and process inertia across key stages such as collection, sorting, dismantling, material extraction, and refurbishment. On top of this, AIbased components—including predictive models and reinforcement learning—are employed to tune process parameters, forecast material yields, and identify optimal routing strategies. The resulting fractionalAI model captures both the physical structure and historical behavior of the recycling system. Numerical experiments show that the proposed approach achieves higher recovery rates, smoother system dynamics, and better energy efficiency compared to classical integerorder and purely datadriven models. The framework offers a mathematically grounded and practically adaptable foundation for designing intelligent, sustainable ewaste recycling operations aligned with circulareconomy principles.