HEURISTICS FOR THE STUDENTS CLUSTERING PROBLEM IN SCHOOLS

Efficient allocation of the students to the nearest schools is crucial, taking into consideration the capacities of the schools and the demands of the students. The problem can be considered as a capacitated clustering problem, in which the schools present different capacities for each grade level. Considering that this class of problems is NP-hard, the proposition of approximate algorithms is of great importance to provide high-quality solutions within an admissible computational effort. This research aims to propose to present three constructive heuristics for the student clustering problem. Computational experiments on a set of 120 randomly generated instances indicate the need for using metaheuristics to solve the problem, as the proposed model failed to return feasible integer solutions within a time limit of 1200 seconds.