School supply-demand balance and accessibility optimisation: A bi-objective spatial matching model using genetic algorithm

A well-balanced spatial matching of schools can reflect the fairness of educational resource allocation. However, problems such as unbalanced allocation, low facility utilisation rate, and excess commuting persist. To optimise the spatial matching between schools and residential areas, this paper proposes a novel Linear and Nonlinear Bi-objective Optimisation Model (LNBOM), which aims to achieve the best school allocation plan by optimising the supply-demand gap and accessibility and adopting an improved Non-dominated Sorting Genetic Algorithm II(NSGA-II). The proposed bi-objective optimisation model represents an extension to the literature on accessibility and excess commuting. The model has already been applied in Nanning, China. The results show that in different demographic scenarios, the LNBOM model can achieve optimised results, significantly reducing the students' overall commuting distance while greatly increasing the utilisation rate and accessibility of schools. Additionally, when the supply-demand gap increases, the model optimises accessibility more effectively. Conversely, the optimisation effect on the supply-demand gap is even better. Furthermore, the model offers several policy implications across various domains, including school districting, the equitable distribution of educational resources, school locational decision-making, and the promotion of sustainable commuting. Although this article takes Chinese schools as a case study, the optimisation model can be applied to other public facilities in different countries under specific supply-demand relationships.