• Journal of KIBIM
• /
• v.14 no.3
• /
• pp.22-30
• /
• 2024

Recently, LLM (Large Language Model), a rapidly developing generative AI technology, is receiving much attention in the smart construction field. This study proposes a methodology for implementing an knowledge expert system by linking BIM (Building Information Modeling), which supports data hub functions in the smart construction domain with LLM. In order to effectively utilize LLM in a BIM expert system, excessive model learning costs, BIM big data processing, and hallucination problems must be solved. This study proposes an LLM-based BIM expert system architecture that considers these problems. This study focuses on the RAG (Retrieval-Augmented Generation) document generation method and search algorithm for effective BIM data retrieval, with the goal of implementing an LLM-based BIM expert system within a small GPU resource. For performance comparison and analysis, a prototype of the designed system is developed, and implications to be considered when developing an LLM-based BIM expert system are derived.

• Korean Journal of Computational Design and Engineering
• /
• v.21 no.3
• /
• pp.252-266
• /
• 2016

An architectural pre-design, which is conducted prior to the architecture design, supports fundamental configuration during the entire AEC project by predicting the cost, demand, etc., of the building, and is therefore gaining importance. In particular, the massing calculation of the pre-design phase should be prioritized, as it is fundamental to architectural outline. However, most architects depend on only their experience and intuition while conceptualizing an integrated framework of design conditions, including the building code and requirements for the massing calculation of the object. Therefore, many difficulties arise in terms of performing appropriate tasks. Thus, the purpose of this study is to implement a knowledge-based BIM for explicitly representing the design knowledge, which is the basis of decision making for an architect while performing the massing calculation. In particular, the 3D knowledge relevant to a project can be provided and accumulated in the massing calculation by the BIM system; this facilitates an integral understanding. Consequently, the approximate result of massing calculation in 3D BIM environment, through both the knowledge-based BIM template and plug-in, can be swiftly provided to the architect. In addition, the architect can invent various alternatives, estimate resulting costs, and reuse the accumulated knowledge in future BIM design processes.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2018.05a
• /
• pp.68-69
• /
• 2018

The construction drawing is a collection of knowledge that has accumulated knowledge and experience of designers and is highly valuable for recycling in the future. BIM is a process and technology that acquires and manages the information needed for the whole life cycle of a building. In order to recycle digitalized 2D drawings in BIM Tool, it is required to use 3D information based on the drawing information to save and reuse them, but there is little research on related methods or system development. In this study, 3D elements are created based on 2D drawing information, and libraries are constructed, and BIM Tool is used to improve the recyclability of drawing information.

• Korean Journal of Computational Design and Engineering
• /
• v.22 no.1
• /
• pp.18-27
• /
• 2017

As building space constitution and layout are critical for satisfying the building owner and users, their optimization is so important in the design process. However it's not always simple to set up objective criteria for the space layout optimization for different requirements and the architects mostly depend on their own experience for these. This study is to suggest a way to make up for this issue by referencing and deducing the space layout based on the given BIM space information and existing knowledge. For this purpose, the Space Syntax is applied to extract the information from a space model and an Expert System is used to make the best use of the relevant knowledge. Based on the Integration indexes for all the spaces, we could compare the space layout alternatives and determine the best selection for different accessibility conditions.

• International conference on construction engineering and project management
• /
• 2015.10a
• /
• pp.60-62
• /
• 2015

The BIM-based architectural design can be the optimal method to improve all sectors of construction industry and reduce environmental impact through digital technologies. The goals and effects of BIM needs in LH to be newly planned based on the experience of public sectors, which will ultimately contribute to maximization of effectiveness. Referring the excellent case in the UK, the current BIM standards according to maturity level 2 handle the contents related to modeling, collaborative data production, data management and assure BIM data have to be used from building construction to operation and maintenance of buildings. Therefore the strategy for the utilization of BIM for construction projects in LH consisted of three steps; first, the minimization of defects via BIM: second, strengthening the competitiveness of construction and asset management: third, upgrading the system of building production and maintenance control. After this trial, LH can accumulate know-how for building construction to be more costeffective and competitive.

• Korean Journal of Computational Design and Engineering
• /
• v.21 no.2
• /
• pp.99-110
• /
• 2016

The quality of building space program is a key to measure how the building performance satisfies its owner and users. However assuring its efficiency requires reliable criteria that reflect high level experience knowledge in the field. This study suggests a plan to gauge the level of building space performance using expert knowledge, which has not been utilized well enough but should play a critical role. In order to setup an expert system measuring level of the space program, we firstly optimized the space areas to the best case in a knowledgebase and use them as criteria to judge the quality of the spaces extracted from BIM model. We found the experimental results show us a promising way of measuring a relative quality of the space programs.

• International conference on construction engineering and project management
• /
• 2013.01a
• /
• pp.426-432
• /
• 2013

The construction industry consists of various and massive architectural information as an architectural process includes a variety of design stages with cooperation of many disciplines. Particularly, architectural information is generated and managed through the life cycle of a building, from conceptual design stage to the construction and maintenance. A Building Information Model (BIM) serves as a shared knowledge resource for information about a facility forming a reliable basis for decisions during its life-cycle from inception onward. BIM technology accomplished quantitative development being utilized in various disciplines. However, it is necessary to develop environment and requirement for qualitative improvement of BIM based project. Particularly, requirement is very important for architectural design evaluations. The purpose of this study is to develop and apply of quality control requirement for improving the quality of architectural design in open BIM environments. To achieve this purpose, the authors have investigated case study for open BIM data quality control (software, guideline and application case) and classified quality control targets according to physical/logical quality and data quality. In addition, the authors have defined open BIM based quality control process and developed quality control requirements. Finally, the authors have developed rule based quality check system using requirements for efficient quality control based on open BIM.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.3
• /
• pp.177-188
• /
• 2013

The construction industry consists of various and massive architectural information as an architectural process includes a variety of design stages with cooperation of many disciplines. Particularly, architectural information is generated and managed through the life cycle of a building, from conceptual design stage to the construction and maintenance. A Building Information Model (BIM) serves as a shared knowledge resource for information about a facility forming a reliable basis for decisions during its life-cycle from inception onward. BIM technology accomplished quantitative development being utilized in various disciplines. However, it is necessary to develop environment and requirement for qualitative improvement of BIM based project. Particularly, requirement is very important for architectural design evaluations. The purpose of this study is to develop and apply of quality control check-list for improving the quality of architectural design in BIM environments. To achieve this purpose, the authors have investigated case study for open BIM data quality control (software, guideline and application case) and classified quality control targets according to physical/logical quality control and data quality. In addition, the authors have defined open BIM based quality control process and developed quality control check-list. Finally, the authors have developed automatic quality check system using requirements for efficient quality control based on open BIM.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.6
• /
• pp.39-46
• /
• 2019

In the 21st century, medical facility projects are required to operate appropriate digital technologies as the development of technology and the interests of various participants become more complex. In order for architects to successfully lead negotiations among various stakeholder groups, it is necessary to plan for effective communication through appropriate design reflecting their opinions and coordination of conflicts. For this purpose, building information modeling (BIM), which is a method of designing based on knowledge information related to medical facilities in the building database, can respond to change of order promptly and minimizes the occurrence of design errors can do. Recently, BIM technology and cloud computing technology in ICT have been combined and research on cloud BIM has been actively carried out. The use of cloud computing technology in BIM-based healthcare facility projects can effectively support decision making among project participants and has the advantage of sharing and collaborating on various forms of information generated during the design process, regardless of location and time. Therefore, the purpose of this study is to build of system that can support the design of medical facility using cloud computing technology in BIM.

• Journal of KIBIM
• /
• v.6 no.1
• /
• pp.9-17
• /
• 2016

Building information modeling refers to combination or set of technologies and organizational solutions that are expected to increase collaboration in the construction industry and to improve the productivity and quality of the design, construction, and maintenance of buildings. For enhanced communication among project participants, various information which BIM model usually includes is provided, furthermore data which contain exchange of unstructured information is needed. If the extension of BIM standard file format for practical use of design Issue information about collaborative design process is fulfilled, the productivity and quality of design will be improved.