Multiobjective Assignment of Citizens to INE Service Modules Using NSGA-II: An Efficient Optimization Approach

Abstract

Efficient allocation of citizens to service modules is crucial for the National Electoral Institute (INE) of Mexico, particularly for those applying for their voter identification card for the first time. This study focuses on a multi-objective optimization approach, using the Second Generation Non-Dominated Genetic Algorithm (NSGA-II), to assign citizens from 16 municipalities in the Toluca Valley to 10 INE service modules. The involved municipalities include Almoloya de Juárez, Calimaya, Chapultepec, Lerma, Metepec, among others, and the service modules are distributed in locations such as Almoloya de Juárez, Metepec, Lerma, and more. Two objective functions are utilized: (1) Maximizing facility coverage, ensuring that the largest number of citizens is assigned to a module, subject to the capacity constraints of each module, and (2) Minimizing transportation costs by reducing the total distance citizens need to travel to reach the modules. The NSGA-II algorithm is compared to alternative optimization methods, and fuzzy logic is employed to evaluate the quality of the generated solutions. The results demonstrate that the NSGA-II-based approach outperforms alternative methods in both efficiency and solution quality, and its impact on INE practice is discussed.