An efficient model for locating solid waste collection sites in urban residential areas

Abstract

A new model for the efficient collection of solid waste in urban residential areas is proposed as a set covering facility location problem. The model finds the optimal location of waste collection sites such that all the customers are covered. An important feature of the model is that, rather than aggregate the quantity of waste from a group of demand points, it is assumed that the volume of waste of different types is known for each customer. This assumption encourages point-of-collection sorting which helps provide an improved solution to the problem of waste collection. A Lagrangian relaxation (LR) of the problem was developed, and the resulting dual problem was solved using an LR procedure in which the vectors of Lagrangian multipliers were updated through the subgradient optimisation. A simple linear relaxation heuristic was developed to obtain a feasible solution to the primal problem. Five newly constructed datasets mimicking the scenario of a local municipality were used to test the efficiency of the model and the solution technique. Preliminary results and comparison with existing results showed that the proposed model is efficient as considerable reductions were obtained for the total number of activated collection sites and allocated containers.