Learning-driven construction productivity prediction for prefabricated external insulation wall system

Jaemin Jeong, Jaewook Jeong, Jaehyun Lee, Daeho Kim, Jeong Wook Son

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

The recent shortage of young skilled laborers is one of the impending issues facing the global construction industry. To address these issues, the prefabricated external insulation system (PEIS) can be suggested as an alternative. However, before applying it to construction projects, a construction productivity analysis is difficult due to the complexity of simulation modeling and the absence of real data. Thus, this paper aims to develop a learning-based productivity prediction model for PEIS using machine learning. This describes a learning-based productivity prediction model for PEIS using machine learning and consists of three steps: (i) Establishment of data, (ii) Development of activity cycle diagram for PEIS, and (iii) Prediction model for productivity analysis. The prediction model has a precision rate of 99.09%. This paper contributes to the literature by developing the possibility of a quick analysis of construction productivity without real data through a machine learning approach.

Original languageEnglish
Article number104441
JournalAutomation in Construction
Volume141
DOIs
StatePublished - Sep 2022

Bibliographical note

Funding Information:
This research was financially supported by the Ministry of Trade, Industry and Energy (MOTIE) and Korea Institute for Advancement of Technology (KIAT) through the International Cooperative R&D program. (Project No. P0017191 ).

Publisher Copyright:
© 2022 Elsevier B.V.

Keywords

  • Activity cycle diagrams
  • Construction productivity
  • K-fold-cross validation
  • Machine learning
  • Prefabricated external insulation wall

Fingerprint

Dive into the research topics of 'Learning-driven construction productivity prediction for prefabricated external insulation wall system'. Together they form a unique fingerprint.

Cite this