Literature research urban logistics has shown us the growing interest in sustainable supply chain serving the population, which must necessarily be based on location and routing of vehicles in order, using the integer linear programming methodology, to model these cases. This study presents a sustainable four-level bi-objective model to optimize economic costs and measure the amount of carbon dioxide generated by the transportation process. The process consists of a provider that supplies a distribution center or warehouse, which also serves a group of facilities located by a group of customers who expect to receive attention to their two orders, one for the purchase of the acquired product and the other for the delivery of the waste generated. Finally, the waste is transported to specialized centers. The model implemented with GLPK obtained optimal results for small and medium-sized instances. For scenarios with more than 40 customers, it was not possible to find solutions, as the computational processing time limit was exceeded by 7 200 s, for these scenarios it is recommended to solve the model using metaheuristics.