The increasing volume of poultry waste and the growing emphasis on resource recovery have made the development of smart waste management systems a strategic priority for achieving circular economy objectives. However, designing such systems remains challenging due to uncertain waste generation, complex facility location decisions, and the need to integrate digital technologies into waste management networks. To address these challenges, this study proposes a comprehensive data-driven optimization framework for designing a smart poultry waste management network based on circular economy principles and Industry 4.0 technologies. The proposed methodology integrates the Fuzzy Best–Worst Method (FBWM) to determine the importance of facility location criteria, the LightGBM machine learning algorithm to evaluate candidate locations, and a multi-objective mathematical programming model to configure the waste management network. Demand uncertainty is addressed through a hybrid data-driven approach combining Seasonal AutoRegressive Integrated Moving Average (SARIMA) forecasting and chance-constrained fuzzy programming, while the resulting optimization model is solved using the Chebyshev Multi-Choice Goal Programming with Utility Function (CMCGP-UF). The proposed framework simultaneously maximizes network profitability, facility quality, service level, and network visibility by incorporating blockchain-enabled information-sharing systems into the network design. The computational results demonstrate that the proposed framework effectively balances multiple strategic objectives while improving resource utilization throughout the network. The findings indicate that integrating machine learning, predictive analytics, circular economy principles, and blockchain-based information sharing significantly enhances facility selection, service performance, and network visibility, providing a robust decision-support framework for developing intelligent and economically efficient poultry waste management systems under uncertainty.