Vision AI–Integrated 4D BIM for Real-Time Construction Progress Monitoring and Anomaly Detection
Civil Engineering Science and Technology · 2026
DOI: 10.51903/rh1m9251 ↗Kurz gefasst
Fachartikel in der Zeitschrift Civil Engineering Science and Technology, 2026. Verfasst von Kieran Ashcrof und 1 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Construction progress monitoring and quality inspection in high-rise building projects are still largely dependent on manual processes, resulting in inefficient reporting, delayed decision-making, and inconsistent inspection quality. Existing studies generally focus on either Vision-AI-based progress monitoring or BIM-based schedule management, with limited integration of geometric anomaly detection into a unified framework. This study aims to develop and evaluate an integrated Vision-AI and 4D-BIM framework for automated construction progress monitoring and structural geometric anomaly detection. The proposed method combines transformer-based object detection, three-dimensional coordinate transformation, and 4D-BIM schedule synchronization to identify structural components, compare as-built conditions with BIM models, and monitor construction progress in real time. Experimental results demonstrated an average precision of 0.882, an average recall of 0.847, and an inference speed of 41.5 FPS, while successfully detecting structural deviations exceeding predefined tolerance limits and synchronizing construction progress with planned schedules. The novelty of this research lies in the integration of transformer-based Vision-AI, 4D-BIM synchronization, and automated geometric anomaly detection within a single computational framework capable of simultaneously performing progress monitoring and dimensional quality assessment. The proposed framework contributes to digital construction management by improving monitoring efficiency, reducing reliance on manual inspection, and supporting faster identification of schedule deviations and quality issues. These findings demonstrate that the proposed framework provides an effective and practical solution for intelligent construction monitoring in high-rise building projects.
Zitieren
APA
Ashcrof, K., & Donovan, M. (2026). Vision AI–Integrated 4D BIM for Real-Time Construction Progress Monitoring and Anomaly Detection. Civil Engineering Science and Technology. https://doi.org/10.51903/rh1m9251
Vancouver
Ashcrof K, Donovan M. Vision AI–Integrated 4D BIM for Real-Time Construction Progress Monitoring and Anomaly Detection. Civil Engineering Science and Technology. 2026. doi:10.51903/rh1m9251
BibTeX
@article{Ashcrof2026,
author = {Ashcrof, Kieran and Donovan, Maeve},
title = {Vision AI–Integrated 4D BIM for Real-Time Construction Progress Monitoring and Anomaly Detection},
journal = {Civil Engineering Science and Technology},
year = {2026},
doi = {10.51903/rh1m9251}
}
EUVIMED