School Zone Mapping Based on Education and Population Data Using Genetic Algorithm

The New Student Admission System through zoning in Indonesia, aimed at distributing students evenly, often leads to discrepancies with demand, resulting in student shortages in some schools. One of the main challenges is the difficulty in determining appropriate zoning boundaries. This research addresses the issue of student shortages in schools and prospective students not getting assigned to schools due to the lack of involvement of education and population data in school zone determination. Genetic Algorithms are proposed as an effective solution to optimize school zoning and address student enrollment imbalances in a region, given their ability to tackle complex problems and find global solutions. The Education and Population data used include the number and capacity of schools, as well as the number of subdistrict and the population aged 6 years old. The population initialization is done by forming chromosomes from pairs of subdistrict and schools. This research utilizes two main fitness functions, namely distance fitness and capacity fitness, to evaluate the effectiveness of school zoning. The advantage of this approach lies in its mathematical consideration of the central coordinates of residential areas, as well as structured and objective calculations. By considering school distance and capacity, this research achieves the best fitness value. Compared to previous school zoning mapping methods, the use of Genetic Algorithms yields better results. This is because the analysis of school capacities shows that all prospective students can be accommodated in public schools, and all schools receive students even though some schools are not fully occupied. Thus, the use of Genetic Algorithms in zoning mapping can be a solution to address student enrollment imbalances in a region.