This paper proposes an IoT-based smart waste management system architecture for Sarajevo, integrating real-time sensor data, predictive analytics, and robust communication protocols. Contributions include a modular, scalable design tailored to Sarajevo’s urban challenges, an approach to a combination of IoT and ML for dynamic waste management, and a framework adaptable to mid-sized cities globally. Such a system, designed to enhance the efficiency and sustainability of waste collection in urban environments, grounds the way for future empirical validation. Utilizing real-time data from sensor-equipped bins, a cloud-based backend, and machine learning algorithms, the system provides real-time monitoring, predictive analytics, and route optimization. Key outcomes include reduced operational costs, optimized fuel consumption, and enhanced urban cleanliness. The modular system architecture and the integration of robust data transmission protocols make it adaptable to similar mid-sized cities globally.