Метод AI-оптимізації функціонального зонування в девелоперських проєктах освітніх середовищ

Traditional management of educational development projects relies heavily on a subjective, experience-based approach to spatial planning, which leads to a limited exploration of design alternatives and a weak connection between initial decisions and long-term lifecycle performance. This linear process lacks a methodical toolkit for navigating the complex trade-offs between cost, functionality, and future adaptability under conditions of high uncertainty. To address these limitations, this study develops and proposes a method for the AI-optimization of functional zoning in educational development projects. This method is based on a structured framework that integrates a Genetic Algorithm with a Multi-Criteria Decision Analysis (MCDA) model by formalizing stakeholder requirements into a set of mathematically verified and comparable project scenarios. The core of the developed method is a formalized algorithmic process that functions as a generative decision support system. The process begins with the digitization of project constraints, including building codes, budget limits, and a weighted adjacency graph representing the topological requirements between functional zones. The generative engine then initializes a population of random layouts and iteratively refines them through the genetic operators of selection, crossover, and mutation. The fitness of each candidate is evaluated using an objective function vector that simultaneously optimizes three conflicting criteria: (1) minimization of Lifecycle Cost (LCC), which includes both capital and operational expenditures; (2) maximization of Functional Utility, measured through student flow efficiency and adjacency compliance; and (3) maximization of Adaptability, assessed by the layout’s modularity and potential for future expansion. The output of the method is not a single solution but a Pareto set, which presents a collection of non-dominated solutions for managerial analysis. The proposed method for AI-optimization of functional zoning marks a paradigm shift from conventional, reactive project management to a proactive, predictive approach. It is anticipated that this method will enhance the effectiveness of decision-making during the pre-investment phase of a development project and provide managers with a reliable, evidence-based foundation for selecting the optimal configuration. The practical significance of the method lies in the generation of a Pareto set, which enables stakeholders to make informed and defensible trade-off decisions among financial, pedagogical, and strategic goals. This enhances the project’s digital resilience, minimizes the risks of scope creep and functional obsolescence, and ultimately ensures that the capital investment creates a sustainable, efficient, and adaptive educational asset.