Urban solid waste management is an increasing global challenge driven by population growth and urbanization. Recycling rates remain below 10% in many developing countries due to inadequate infrastructure and planning. This paper proposes a bi-level programming model to optimize urban solid waste collection and management in a hierarchical supply chain involving transfer stations and specialized treatment facilities. The model captures the decision-making processes of the government, responsible for determining the location of transfer stations and managing waste transportation, and a private company in charge of waste recycling, plant specialization. The upper-level objective, pursued by the government, is to minimize carbon dioxide (CO2) emissions across the supply chain, while the lower-level objective, driven by the private company, seeks to reduce shipping and operational costs. A novel matheuristic algorithm based on a nested local search is developed to solve the bi-level model. The proposed matheuristic incorporates a deconstructive process to generate upper-level solutions. Moreover, bi-level feasibility is ensured by optimally solving the parameterized lower-level problem. Extensive computational experiments were conducted, covering two real-world case studies: Medellín, Colombia, and Santiago, Chile. A budget sensitivity analysis was performed, providing valuable managerial insights into the economic-environmental trade-offs in waste management planning. Furthermore, large-scale instances were tested to evaluate the scalability of the proposed matheuristic. An analysis excluding the private company¿s decisions highlighted the advantages of the bi-level approach. The results underscore the potential of the bi-level framework to support strategic decision-making in urban waste management systems, particularly when a hierarchy exists among decision-makers.
