On-Site Construction Worker Activity Monitoring Using Deep Learning.
Saved in:
| Title: | On-Site Construction Worker Activity Monitoring Using Deep Learning. |
|---|---|
| Authors: | Monfared, Ehsan1 (AUTHOR) es.monfared@email.kntu.ac.ir, Alipouri, Yaghoub2 (AUTHOR) y.alipouri@kntu.ac.ir |
| Source: | Journal of Computing in Civil Engineering. Jul2026, Vol. 40 Issue 4, p1-20. 20p. |
| Subjects: | Human activity recognition, Wearable technology, Labor productivity, Deep learning, Employee surveillance, Construction management, Convolutional neural networks, Recurrent neural networks |
| Abstract: | Low productivity in the construction industry compared with other sectors is a longstanding concern. Traditionally, managers have relied on manual sampling of worker activities by monitoring task types and durations to identify and address productivity obstacles. However, this method is labor-intensive, error-prone, and inadequate for the dynamic demands of site management. Previous studies have highlighted the potential of utilizing wristbands equipped with inertial measurement units (IMUs) to automate activity recognition and sampling processes; however, most of these investigations have been confined to controlled laboratory settings. In this study, several commonly performed construction tasks, including those extensively researched, such as tiling, masonry, and painting, as well as less frequently studied activities like manual excavation and wall demolition, were examined under real-world construction site conditions. To enhance practical relevance for site managers, workers' activities are categorized at two levels: first, based on their overall impact on productivity using the traditional activity sampling taxonomy, which classifies activities into direct work, indirect work, and ineffective activities; and second, by their specific activity context. A variety of hybrid algorithms that integrate convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for activity recognition were employed. Among the developed models, the hybrid convolutional neural network combined with a bidirectional gated recurrent unit (CNN-BiGRU) demonstrated the highest performance, achieving classification accuracies of 84.0% at the first level and 79.5% at the second level. These findings validate the effectiveness of using a single wristband equipped with a gyroscope and accelerometer on active construction sites. However, classification accuracy was unsatisfactory for subcategories of indirect work. This study discusses the limitations of implementing such a structure in actual construction sites, offers suggestions that provide new insights for future research, and provides the groundwork for broader practical implementation. Practical Applications: Enhancing labor productivity is a significant challenge in the construction industry, with direct implications for project costs and schedules. Traditional monitoring methods, such as manual observation, are often impractical due to being slow, costly, and disruptive. This research presents a practical, automated alternative using a single, low-cost wristband equipped with motion sensors. By applying deep learning models to the sensor data, the system can automatically classify worker activities throughout the day. This study demonstrates the effectiveness of the approach in distinguishing between value-adding direct work (e.g., painting, masonry), supportive indirect work (e.g., moving materials), and ineffective time on active construction sites. This framework provides construction managers and contractors with an objective real-time tool for gaining insights into on-site operations. By automatically tracking activities, managers can identify productivity bottlenecks, optimize resource allocation, and make data-driven decisions to reduce delays and control costs, ultimately leading to more efficient projects. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Computing in Civil Engineering is the property of American Society of Civil Engineers and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 193805615 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: On-Site Construction Worker Activity Monitoring Using Deep Learning. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Monfared%2C+Ehsan%22">Monfared, Ehsan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> es.monfared@email.kntu.ac.ir</i><br /><searchLink fieldCode="AR" term="%22Alipouri%2C+Yaghoub%22">Alipouri, Yaghoub</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> y.alipouri@kntu.ac.ir</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Journal+of+Computing+in+Civil+Engineering%22">Journal of Computing in Civil Engineering</searchLink>. Jul2026, Vol. 40 Issue 4, p1-20. 20p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Human+activity+recognition%22">Human activity recognition</searchLink><br /><searchLink fieldCode="DE" term="%22Wearable+technology%22">Wearable technology</searchLink><br /><searchLink fieldCode="DE" term="%22Labor+productivity%22">Labor productivity</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink><br /><searchLink fieldCode="DE" term="%22Employee+surveillance%22">Employee surveillance</searchLink><br /><searchLink fieldCode="DE" term="%22Construction+management%22">Construction management</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Recurrent+neural+networks%22">Recurrent neural networks</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Low productivity in the construction industry compared with other sectors is a longstanding concern. Traditionally, managers have relied on manual sampling of worker activities by monitoring task types and durations to identify and address productivity obstacles. However, this method is labor-intensive, error-prone, and inadequate for the dynamic demands of site management. Previous studies have highlighted the potential of utilizing wristbands equipped with inertial measurement units (IMUs) to automate activity recognition and sampling processes; however, most of these investigations have been confined to controlled laboratory settings. In this study, several commonly performed construction tasks, including those extensively researched, such as tiling, masonry, and painting, as well as less frequently studied activities like manual excavation and wall demolition, were examined under real-world construction site conditions. To enhance practical relevance for site managers, workers' activities are categorized at two levels: first, based on their overall impact on productivity using the traditional activity sampling taxonomy, which classifies activities into direct work, indirect work, and ineffective activities; and second, by their specific activity context. A variety of hybrid algorithms that integrate convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for activity recognition were employed. Among the developed models, the hybrid convolutional neural network combined with a bidirectional gated recurrent unit (CNN-BiGRU) demonstrated the highest performance, achieving classification accuracies of 84.0% at the first level and 79.5% at the second level. These findings validate the effectiveness of using a single wristband equipped with a gyroscope and accelerometer on active construction sites. However, classification accuracy was unsatisfactory for subcategories of indirect work. This study discusses the limitations of implementing such a structure in actual construction sites, offers suggestions that provide new insights for future research, and provides the groundwork for broader practical implementation. Practical Applications: Enhancing labor productivity is a significant challenge in the construction industry, with direct implications for project costs and schedules. Traditional monitoring methods, such as manual observation, are often impractical due to being slow, costly, and disruptive. This research presents a practical, automated alternative using a single, low-cost wristband equipped with motion sensors. By applying deep learning models to the sensor data, the system can automatically classify worker activities throughout the day. This study demonstrates the effectiveness of the approach in distinguishing between value-adding direct work (e.g., painting, masonry), supportive indirect work (e.g., moving materials), and ineffective time on active construction sites. This framework provides construction managers and contractors with an objective real-time tool for gaining insights into on-site operations. By automatically tracking activities, managers can identify productivity bottlenecks, optimize resource allocation, and make data-driven decisions to reduce delays and control costs, ultimately leading to more efficient projects. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Journal of Computing in Civil Engineering is the property of American Society of Civil Engineers and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=193805615 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1061/JCCEE5.CPENG-7234 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 20 StartPage: 1 Subjects: – SubjectFull: Human activity recognition Type: general – SubjectFull: Wearable technology Type: general – SubjectFull: Labor productivity Type: general – SubjectFull: Deep learning Type: general – SubjectFull: Employee surveillance Type: general – SubjectFull: Construction management Type: general – SubjectFull: Convolutional neural networks Type: general – SubjectFull: Recurrent neural networks Type: general Titles: – TitleFull: On-Site Construction Worker Activity Monitoring Using Deep Learning. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Monfared, Ehsan – PersonEntity: Name: NameFull: Alipouri, Yaghoub IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 07 Text: Jul2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 08873801 Numbering: – Type: volume Value: 40 – Type: issue Value: 4 Titles: – TitleFull: Journal of Computing in Civil Engineering Type: main |
| ResultId | 1 |
