Publicação
Smart BIM objects for intelligent modular construction
| Resumo: | From a general perspective, AEC is one of the largest industries and surprisingly, when compared to other industries such as the automobile and airplane production, it shows one of the lowest productivity rates. One of the main solutions for increasing productivity rate is to shift the production processes from traditional to manufacturing processes. But there is a problem: Each building is a unique product and the usual manufacturing approach – Mass Production – cannot be directly adapted to the AEC industry. Thanks to new technologies, a new method has emerged: ETO or Engineered-To-Order. In this approach, the products can be fabricated corresponding to the design of the final product (building). An important challenge should be solved before: in ETO approach, the properties of the elements should be extracted from design documents. Hopefully, this problem can be solved with BIM and Artificial Intelligence. Building Information Modelling (BIM) takes advantage of powerful digital processes and new technologies in programming, such as Object-Oriented Programming languages (OOP). With BIM, it is possible to create a complete 3D model enriched by various non-graphic information of the final product. Associating BIM to Engineered-To-Order method makes the overall processes smoother and easier. In addition, there are other technologies embedded in BIM that make processes completely automatic: Parametric Design and Artificial Intelligence that led to Smart BIM Objects. By applying these technologies, designing and modelling elements can be done automatically in an optimized way. Within the current context, as the AEC industry increasingly takes advantage of intelligence adaptable digital representative of building elements. This work tries to implement AI in the work of panelling floors automatically. The research focuses on exploring how BIM capabilities and Artificial intelligence can support the production processes, in the ETO approach. To be more precise, this dissertation tries to utilize the BIM environment to make the process of panelling a floor or a wall with openings automatic. In ETO production method, it is very difficult to find a balance between the design freedom and the production limits. How is it possible to integrate the limits of production process into the design process? There are various approaches but the one used in this research, is Artificial Intelligence. The challenge is to design panels to be applied in building elements (wall, floor, and roof), considering the production rules and some restrictions. This research tries to associate the shape of elements with the production process of panels while considering the designer’s desires and the fabricator’s rules. Therefore, an algorithm was developed to create ETO panels fitted to construction elements that the designer created for the building, considering the various rules of production process. The algorithm collects the shape of element, starts to divide it to some smaller parts appropriate for panelling, modifies and finalize them for panelling. Then starts to arrange the panels and defines the size of them in an optimised way responding to production rules and limits. The algorithm developed uses the Python programming interface of Grasshopper and for the inputs and outputs, uses Tekla Structures as a BIM authoring software. This automated process not only reduces the time of work and resulting costs, but also produces more precise and optimized results. This algorithm is capable to incorporate rules that are defined by the user and calculates the result that correspond to those rules. |
|---|---|
| Autores principais: | Parsa, Mohsen |
| Assunto: | BIM Smart BIM objects Engineered to order (ETO) Pre-fabricated panel Python Grasshopper Tekla structures Painel Pré-fabricado Engenharia e Tecnologia::Engenharia Civil |
| Ano: | 2021 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | From a general perspective, AEC is one of the largest industries and surprisingly, when compared to other industries such as the automobile and airplane production, it shows one of the lowest productivity rates. One of the main solutions for increasing productivity rate is to shift the production processes from traditional to manufacturing processes. But there is a problem: Each building is a unique product and the usual manufacturing approach – Mass Production – cannot be directly adapted to the AEC industry. Thanks to new technologies, a new method has emerged: ETO or Engineered-To-Order. In this approach, the products can be fabricated corresponding to the design of the final product (building). An important challenge should be solved before: in ETO approach, the properties of the elements should be extracted from design documents. Hopefully, this problem can be solved with BIM and Artificial Intelligence. Building Information Modelling (BIM) takes advantage of powerful digital processes and new technologies in programming, such as Object-Oriented Programming languages (OOP). With BIM, it is possible to create a complete 3D model enriched by various non-graphic information of the final product. Associating BIM to Engineered-To-Order method makes the overall processes smoother and easier. In addition, there are other technologies embedded in BIM that make processes completely automatic: Parametric Design and Artificial Intelligence that led to Smart BIM Objects. By applying these technologies, designing and modelling elements can be done automatically in an optimized way. Within the current context, as the AEC industry increasingly takes advantage of intelligence adaptable digital representative of building elements. This work tries to implement AI in the work of panelling floors automatically. The research focuses on exploring how BIM capabilities and Artificial intelligence can support the production processes, in the ETO approach. To be more precise, this dissertation tries to utilize the BIM environment to make the process of panelling a floor or a wall with openings automatic. In ETO production method, it is very difficult to find a balance between the design freedom and the production limits. How is it possible to integrate the limits of production process into the design process? There are various approaches but the one used in this research, is Artificial Intelligence. The challenge is to design panels to be applied in building elements (wall, floor, and roof), considering the production rules and some restrictions. This research tries to associate the shape of elements with the production process of panels while considering the designer’s desires and the fabricator’s rules. Therefore, an algorithm was developed to create ETO panels fitted to construction elements that the designer created for the building, considering the various rules of production process. The algorithm collects the shape of element, starts to divide it to some smaller parts appropriate for panelling, modifies and finalize them for panelling. Then starts to arrange the panels and defines the size of them in an optimised way responding to production rules and limits. The algorithm developed uses the Python programming interface of Grasshopper and for the inputs and outputs, uses Tekla Structures as a BIM authoring software. This automated process not only reduces the time of work and resulting costs, but also produces more precise and optimized results. This algorithm is capable to incorporate rules that are defined by the user and calculates the result that correspond to those rules. |
|---|