DEVELOPMENT OF A DEEP LEARNING-POWERED SYSTEM FOR CLASSIFYING CONSTRUCTION WORKER PRODUCTIVITY PATTERNS
Keywords:
Automated System, Deep Learning, Productivity BehaviorAbstract
In the construction industry, worker productivity is crucial for project success, but it is often challenging to accurately assess and monitor. Traditional manual methods of evaluating productivity can be time-consuming and subjective. This study presents the development of an automated system for classifying the productivity behavior of construction workers using deep learning. By employing computer vision and deep learning techniques, the system captures real-time video footage of workers and classifies their behavior into different productivity categories. The system is trained using a large dataset of labeled images representing various worker actions. The results indicate that the system is capable of accurately classifying worker behavior with high precision, offering a more efficient and objective way of assessing productivity on construction sites.
Downloads
References
S. Kusayanagi, “A study on productivity improvement program for international construction projects,” Japanese Journal of JSCE, no. 528/VI-29, pp. 143-154, 1995.
Y. Gong, K. Yang, J. O. Seo, and J. G. Lee, “Wearable acceleration-based action recognition for long-term and continuous activity analysis in construction site,” Journal of Building Engineering, vol. 52, 2022.
M. Kusumoto, A. Aini, and C. Pang-Jo, “Case study of artificial intelligence at construction sites,” Artificial Intelligence and Data Science, vol. 1, no. J1, pp. 301-306, Nov. 2020.
Takenaka Corporation, “System to track the location of elevating vehicles and workers in the construction site,” Available: https://www.takenaka.co.jp/news/2016/02/01/index.html. Accessed: May 10, 2022.
T. Goso, A. Ochi, and S. Kusayanagi, Journal of Infrastructure Planning and Management, vol. 66, no. 1, pp. 317-328, 2010.
T. Goso and T. Hiromitsu, “Productivity management data collection and analysis system using ZigBee and acceleration sensors at construction sites,” Monthly Automatic Recognition, vol. 8, pp. 25-30, 2015.
T. Goso, “Development and improvement of labor productivity assessment system in construction site by using information technology,” Journal of Society for Social Management Systems, pp. SMS11-9178, 2012.
S. Kusayanagi and T. Goso, “Visualization in the construction management field—Visualizing the project execution process,” JSCE Magazine, Civil Engineering, vol. 96, pp. 30-32, 2011
T. Osawa and T. Goso, “Development of an automated program for work identification at construction sites,” in Proceedings of the 75th Annual Conference of the Japan Society of Civil Engineers, 2020
T. Osawa and T. Goso, “Construction of a productivity classification system based on a combination of pre-classification and machine learning,” in Proceedings of the Symposium on Civil Engineering Informatics, Sept. 2021, vol. 46, pp. 1-4.
T. Goso, K. Muto, M. Hamano, T. Osawa, and R. Kasai, “Development of a system for collecting and analyzing labor environment and productivity data from construction sites using machine learning,” Journal of Pavement Technology, vol. 57, no. 6, pp. 3-8, 2022.
N. Hashimoto and S. Takamaeda, “Investigation of approximate computation methods for machine learning-based video processing,” IEICE Tech. Rep., vol. CPSY2021-59, no. DC2021-93, 2022.
M. H. Uddin, J. M. Kanon Ara, M. H. Rahman, and S. H. Yang, “A study of real-time physical activity recognition from motion sensors via smartphone using deep neural network,” in Proc. 5th International Conference on Electrical Information and Communication Technology (EICT), 2021.
K. Lee and S. Han, “Convolutional neural network modeling strategy for fall-related motion recognition using acceleration features of a scaffolding structure,” Automation in Construction, vol. 130, Oct. 2021.
R. Imai, D. Kamiya, H. Inoue, S. Tanaka, J. Sakurai, K. Sakamoto, T. Fujii, K. Mimura, and M. Ito, “Study on trial of person identification for safety management at construction sites,” in Proceedings of the Symposium on Civil Engineering Informatics, 2018, vol. 43, no. 37, pp. 145-148.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Kobayashi Tanaka
This work is licensed under a Creative Commons Attribution 4.0 International License.
All content published in the Journal of Applied Science and Social Science (JASSS) is protected by copyright. Authors retain the copyright to their work, and grant JASSS the right to publish the work under a Creative Commons Attribution License (CC BY). This license allows others to distribute, remix, adapt, and build upon the work, even commercially, as long as they credit the author(s) for the original creation.
