Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Building and Real Estate | en_US |
| dc.contributor.advisor | Seo, JoonOh (BRE) | en_US |
| dc.creator | Gong, Yue | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/12919 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Automated construction worker productivity assessment using sensor fusion approaches | en_US |
| dcterms.abstract | The construction industry is among the most labor-intensive sectors, with many tasks dependent on manual labor. This heavy reliance on a manual workforce presents challenges in managing ongoing projects, which is recognized as a fundamental cause of low productivity in construction. Therefore, identifying the worker-related root causes through continuous monitoring and assessment of activities is essential to addressing the low productivity issue. However, continuous activity monitoring for productivity analysis at the construction sites is challenging due to the environmental complexity and the dynamic nature of construction projects. Although observational methods are widely used to collect activity-related data, such as the locations of tasks performed and types of activities, these methods face criticism for their time-consuming and manual nature of data collection. For efficient monitoring of individual workers, previous research efforts have argued the need for automated approaches for field data collection by using sensing technologies, including cameras and wearable sensors. | en_US |
| dcterms.abstract | Previous studies have proposed various sensor-based approaches for monitoring workers’ activities. Nonetheless, significant challenges persist in identifying productivity issues. Firstly, most research in sensor-based activity monitoring categorizes activities based on repetitive tasks. However, the unstandardized nature of construction work means these predefined work taxonomies are not universally applicable, failing to recognize different working contexts essential for identifying core productivity problems. Secondly, existing sensor-based methods have primarily been validated in controlled environments, leaving the efficacy of these approaches for long-term, continuous activity data collection untested in field conditions. Thirdly, existing studies often depend on a single sensor data source, demonstrating acceptable accuracy in detecting various construction activities. However, each sensor has inherent strengths and weaknesses, and reliance on a single data source could result in significant errors, particularly in challenging environments. | en_US |
| dcterms.abstract | The current study aims to develop a comprehensive sensor-fusion-based automated activity assessment framework to identify potential worker-related productivity issues. The designed framework involves continuously collecting activity data using multi-modal sensors, including Bluetooth Low Energy (BLE) beacons for location tracking and accelerometers and cameras for activity monitoring. Specifically, this study established three objectives to address the research challenges outlined in the previous paragraph. The first objective is to design a refined taxonomy for construction activities, enhancing worker monitoring accuracy with work context information. The second objective is to assess the effectiveness of BLE beacon-based location tracking and accelerometer-based activity monitoring in diverse field settings. The third objective is to develop and evaluate a sensor fusion method that combines accelerometer and video data to improve activity recognition robustness in construction environments. The proposed worker activity assessment framework is expected to collect activity and location information from construction workers in real-time, aiding in better understanding individual worker-level productivity issues and determining the most suitable intervention strategies to improve construction productivity. | en_US |
| dcterms.extent | xv, 174 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2024 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.LCSH | Construction workers | en_US |
| dcterms.LCSH | Detectors | en_US |
| dcterms.LCSH | Electronic monitoring in the workplace | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12919