Development of Automated System for Classifying Productivity Behavior of Construction Workers Using Deep Learning

Authors

  • Ryuji Kasai Tokyo City University
  • Takashi Goso Tokyo City University
  • Mizuki Akiyama Tokyo City University
  • Hiroto Kato Tokyo City University

DOI:

https://doi.org/10.4186/ej.2024.28.10.59

Keywords:

worker productivity data, work sampling, deep learning, acceleration, convolutional neural network

Abstract

In Japan, the integration and comprehensive understanding of data related to the working environment and productivity at construction sites remain underdeveloped. This study introduces a system that utilizes the human activity recognition method, employing accelerometers combined with deep learning techniques, to capture a detailed overview of activities performed by construction site workers. We developed a new approach for transforming accelerometer data collected from devices attached to workers’ helmets into a format suitable for image-based analysis. This data was then processed using a convolutional neural network to create a deep-learning model capable of distinguishing between different types of worker movements. The model demonstrated high accuracy, with correct classification rates of 80.0% for walking and 92.1% for forward-leaning postures—activities commonly observed at construction sites. Additionally, we established an ensemble system to enhance the final classification of productive motions. This innovative system holds the promise of enabling future quantification of on-site productivity through daily indices that reflect workers’ engagement levels.

Downloads

Download data is not yet available.

Author Biographies

Ryuji Kasai

Tokyo City University, Tokyo, Japan

Takashi Goso

Faculty of Architecture and Urban Design, Tokyo City University, Tokyo, Japan

Mizuki Akiyama

Tokyo City University, Tokyo, Japan

Hiroto Kato

Tokyo City University, Tokyo, Japan

Downloads

Published In
Vol 28 No 10, Oct 31, 2024
How to Cite
[1]
R. Kasai, T. Goso, M. Akiyama, and H. Kato, “Development of Automated System for Classifying Productivity Behavior of Construction Workers Using Deep Learning”, Eng. J., vol. 28, no. 10, pp. 59-76, Oct. 2024.