Impact of artificial neural networks on architectural conceptual creation

• Autores: Acosta Caipa, Karina, Espinoza Molina , Jorge
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad Privada de Tacna
• Repositorio: Revistas - Universidad Privada de Tacna
• Lenguaje: español
• OAI Identifier: oai:revistas.upt.edu.pe:article/1042
• Enlace del recurso: https://revistas.upt.edu.pe/ojs/index.php/arquitek/article/view/1042
• Nivel de acceso: acceso abierto
• Materia: Metodología; espacio; arquitectura; methodology; space; architecture

This study aims to evaluate the incorporation of artificial neural networks in the initial stage of applied architectural design for first-year students. The research, of an applied type with a mixed approach, seeks to determine how this technology facilitates the creation and graphic expression of ideas, improving skills in a short time. The methodology includes two phases: a traditional one, based on manual sketches, and another assisted by AI, comparing the performance in both approaches. The results show that students reduced the time of idea development by 30-40% when using AI and experienced greater satisfaction (98%) when visualizing their concepts in real time. In addition, AI allowed generating more innovative proposals aligned with the project requirements. However, it is noted that it is important to maintain a balance between the use of digital tools and the development of critical and reflective skills. The study concludes that artificial neural networks are an effective complement in the educational process of architectural design.