Large language model-enabled quality reporting for offsite modular housing production

  • Liupengfei Wu
  • , Jinfeng Lou
  • , Liang Yuan
  • , Vikrom Laovisutthichai
  • , Patrick Fong
  • , Xi Chen Chen
  • , Linna Geng

    Research output: Chapter in Book / Conference PaperChapterpeer-review

    Abstract

    The construction sector faces continuous difficulties related to labour shortages, the ageing of its workforce and excessive waste production. Modular Integrated Construction (MiC) provides a beneficial alternative to traditional cast-in-situ procedures by delivering improved productivity levels along with safer working conditions and less material waste. MiC creates integrated modules through manufacturing processes at controlled offsite or offshore facilities which are then transported to assembly locations. The implementation of advanced project management systems such as Enterprise Resource Planning systems and Building Information Modelling and Geographic Information Systems has not fully resolved the critical challenge of efficient quality reporting. This chapter presents a new LLM-based method to improve quality reporting for offsite modular housing production. The method utilizes LLMs to resolve persistent documentation and control problems in quality management while improving MiC production process reliability and efficiency. The results of this research will help develop intelligent and automated quality assurance systems that facilitate wider use of MiC methods.
    Original languageEnglish
    Title of host publicationIntegrated Perspectives in Offsite Construction
    EditorsVikrom Laovisutthichai
    Place of PublicationU.K.
    PublisherRoutlege
    Chapter6
    Pages121-136
    Number of pages16
    ISBN (Electronic)9781003605928
    ISBN (Print)9781032991665
    DOIs
    Publication statusPublished - 2026

    Fingerprint

    Dive into the research topics of 'Large language model-enabled quality reporting for offsite modular housing production'. Together they form a unique fingerprint.

    Cite this