Detail publikace

Wide adoption of AI in Construction: A Case study on machine learning for delay prediction and schedule enhancement with external factors

VENKRBEC, V.

Originální název

Wide adoption of AI in Construction: A Case study on machine learning for delay prediction and schedule enhancement with external factors

Typ

článek v časopise - ostatní, Jost

Jazyk

angličtina

Originální abstrakt

Construction industry frequently faces delays due to both internal and external factors, impacting project schedules and costs. This study explores the use of machine learning (ML) algorithms to predict delays and optimize construction schedules. By integrating various factors, including workforce availability, material delivery timelines, weather conditions, and geopolitical events, the algorithm estimates potential delays in construction projects. Data is extracted from Building Information Modeling (BIM) systems using Industry Foundation Classes (IFC), ensuring a comprehensive and up-to-date dataset. The model leverages regression techniques like XGBoost to train on historical data, enabling accurate predictions of project delays. The results show that machine learning can significantly improve delay prediction accuracy compared to traditional methods, offering a more proactive approach to project management. The integration of external factors such as economic instability and geopolitical tensions further enhances the model’s precision, making it a valuable tool for risk management in modern construction projects.

Klíčová slova

Machine Learning; Construction Delays; BIM; IFC; Schedule Optimization; External Factors.

Autoři

VENKRBEC, V.

Vydáno

31. 12. 2024

Nakladatel

Czech Journal of Civil Engineering

Místo

Brno

ISSN

2336-7148

Periodikum

Czech Journal of Civil Engineering

Ročník

2024

Číslo

01

Stát

Česká republika

Strany od

91

Strany do

103

Strany počet

13

URL

BibTex

@article{BUT196504,
  author="Václav {Venkrbec}",
  title="Wide adoption of AI in Construction: A Case study on machine learning for delay prediction and schedule enhancement with external factors",
  journal="Czech Journal of Civil Engineering",
  year="2024",
  volume="2024",
  number="01",
  pages="91--103",
  doi="10.51704/cjce.2024.vol10.iss1.pp91-103",
  issn="2336-7148",
  url="https://cjce.cz/journal/article/view/231"
}