This study evaluates and validates how BIM functionalities can foster the consideration of lean construction and sustainability principles in building design. To this end, a plugin was developed and implemented in a BIM-enabled visualization environment. This system aims to help a user decide the best design between alternatives, while considering the user’s perspective on what generates value in the project. To illustrate and validate the usability and importance of the system, the design of a warehouse with specific requirements was considered. Four design alternatives were modeled considering different types of superstructures, building envelopes, and materials. The system imported an external database and calculated the general costs, CO₂ emissions, and thermal comfort indicators for each model. It also incorporated user-defined weights for each indicator and graphically displayed the results indicating the best options through a multiple attribute decision method. From the obtained results, it was concluded that it is possible to integrate BIM and lean principles to support sustainable development in the construction industry. Through an experiment with warehouse alternatives, the proposed approach revealed that the best project could save nearly 50% of CO₂ emissions when compared to the worst project and that the second-best project had the best performance in terms of thermal heat flow (1.80 Megawatts). The results also indicate that the system can be a great asset in supporting the decision-making process of selecting the best design alternative and understanding the impact of design changes on cost and on the environment.

A aplicação de Grafos de Conhecimento (KGs) tem ganhado destaque na implementação de soluções em diversas áreas. Na indústria de Óleo e Gás, seu uso é de interesse devido à heterogeneidade dos dados presentes nos softwares de apoio a plantas industrias. Como recursos de padronização desta industria, destaca-se Bibliotecas de dados de Referências disponibilizadas por organizações internacionais e o uso da Ontologia de Dados Industriais (IDO). No entanto, aplicar esses padrões exige estender a IDO com o mapeamento de conceitos específicos do domínio. Neste contexto, este trabalho propõe o estudo sobre o uso de Modelos de Linguagem de Grande Escala (LLMs) como assistente auxiliando a compreensão e sugerindo mapeamentos semânticos.