Architectural Design Using Adaptive Topological Models with Machine Learning: Generative project interrelationships assisted by AI during the design process

Abstract

This article proposes a conceptual and methodological framework for the development of adaptive architectural design systems based on Topological Informational Models (TIM) integrated with artificial learning algorithms. Unlike conventional models that categorize the project according to discrete constructive elements, this approach formalizes the internal and contextual relationships of the project as a relational system. Such a system is capable of acquiring behavior through exposure to variable environmental data, reacting to contextual changes, and suggesting transformations without compromising the project’s relational integrity. The model is validated through simulations of critical scenarios, demonstrating its capacity to maintain the essential design logic while adapting its morphological, functional, or structural configuration.