An Allocation-Routing Problem in Waste Management Planning: Exact and Heuristic Resolution Approaches

Efficient management of municipal solid waste systems is vital for cities today, as it encompasses significant environmental, social, and economic implications. To manage this intricate system, the integration of computational tools to support decision making process is invaluable. This study presents a comprehensive model to simultaneously address two distinct logistics problems, namely the sizing of collection points and the design of collection routes, which are usually treated separately due to their individual complexity. In addition, the model is solved with both a Mixed-Integer Linear Programming (MILP) model and a Genetic Algorithm (GA). Regarding the GA, two chromosomes with two different encoding schemes (binary and permutation) were considered to define the representation and a \(4\times 3^3\) factorial design was performed to tune four of the algorithm parameters: crossover and mutation operators, and their respective rates. Initial results on small instances suggest the potential of GA to address this problem in real-world instances when exact mathematical model are not applicable, since it is able to improve the results obtained with the MILP model.