This chapter reviews the feasibility of employing Large Language Models (LLMs) within the BIM framework for enhanced efficiencies during design, construction, and even facilities management with a focus on LLM characteristics of usability, automation, and communication. LLMs integrate into the Building Information Modeling (BIM) processes to enhance the usability and interoperability of BIM information. LLMs can understand different types of information across diverse platforms and reformat it into usable information. This conversion not only minimizes data loss or error but also fosters effective communication and translatability across various modalities. Furthermore, usability is enhanced on multiple levels; many options taken by the LLM from reporting error findings to presenting ranges of construction parameters serve to simplify processes and avoid unnecessary costs so that project managers can funnel their time into strategic tactical efforts rather than time-consuming menial tasks. Team members similarly stand to benefit as LLMs answer inquiry-based contexts as opposed to general inquiry searches, which leads to the opportunity for more in-depth reporting and summary complexities for greater understanding and agreement. Likewise, the ability of LLMs to generate designs is acknowledged; simply putting parameters and limitations into the system can yield varying options for review by the architect/engineer, which minimizes design time but offers creative yet suitable solutions. However, challenges associated with blended LLMs and BIM systems arise from data quality, training, extensive computing power prerequisites, and ethical concerns surrounding data privacy and accountability. Ultimately, however, LLMs possess the ability to increase interaction with and usability of the data within BIM systems, improve automation processes, and enhance intra- and inter-team communication and generative efforts. Further studies must be conducted surrounding trained data quality, computational potentials, and ethical facets of Artificial Intelligence (AI) in BIM to bring this potential to fruition. If these obstacles can be overcome, never before has the Architectural, Engineering, and Construction (AEC) industry had a potential evolution of integration such that LLMs can foster creativity and productivity during the design/construction process for a more technologically advanced and efficient sustainable future.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Integration of Large Language Model (LLM) and Building Information Modeling (BIM) for Enhanced Construction Project Lifecycle Management: A Review

  • Mallikarjuna Paramesha,
  • Nitin Liladhar Rane,
  • Jayesh Rane

摘要

This chapter reviews the feasibility of employing Large Language Models (LLMs) within the BIM framework for enhanced efficiencies during design, construction, and even facilities management with a focus on LLM characteristics of usability, automation, and communication. LLMs integrate into the Building Information Modeling (BIM) processes to enhance the usability and interoperability of BIM information. LLMs can understand different types of information across diverse platforms and reformat it into usable information. This conversion not only minimizes data loss or error but also fosters effective communication and translatability across various modalities. Furthermore, usability is enhanced on multiple levels; many options taken by the LLM from reporting error findings to presenting ranges of construction parameters serve to simplify processes and avoid unnecessary costs so that project managers can funnel their time into strategic tactical efforts rather than time-consuming menial tasks. Team members similarly stand to benefit as LLMs answer inquiry-based contexts as opposed to general inquiry searches, which leads to the opportunity for more in-depth reporting and summary complexities for greater understanding and agreement. Likewise, the ability of LLMs to generate designs is acknowledged; simply putting parameters and limitations into the system can yield varying options for review by the architect/engineer, which minimizes design time but offers creative yet suitable solutions. However, challenges associated with blended LLMs and BIM systems arise from data quality, training, extensive computing power prerequisites, and ethical concerns surrounding data privacy and accountability. Ultimately, however, LLMs possess the ability to increase interaction with and usability of the data within BIM systems, improve automation processes, and enhance intra- and inter-team communication and generative efforts. Further studies must be conducted surrounding trained data quality, computational potentials, and ethical facets of Artificial Intelligence (AI) in BIM to bring this potential to fruition. If these obstacles can be overcome, never before has the Architectural, Engineering, and Construction (AEC) industry had a potential evolution of integration such that LLMs can foster creativity and productivity during the design/construction process for a more technologically advanced and efficient sustainable future.