Related Experiment Video
Updated: Sep 17, 2026

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
Published on: December 11, 2015
Ergo-PAS: Validation of a smartphone-based real-time posture warning system for ergonomic risk intervention in
Wyke Kusmasari1, Khoirul Muslim1, Yassierli Yassierli1
1Department of Industrial Engineering, Institut Teknologi Bandung, Bandung, Indonesia.
Abstract:
BackgroundConstruction workers are highly exposed to musculoskeletal disorders (MSDs) owing to awkward and repetitive postures. Traditional ergonomic tools lack real-time feedback, and motion-capture systems are impractical for field use. Smartphone-based inertial measurement units (IMUs) offer a cost-effective and portable alternative to traditional methods.ObjectiveThis study aimed to develop and validate the Ergo Posture Analysis System (Ergo-PAS), an Android-based smartphone prototype designed to detect and alert users to high-risk postures in real-time.MethodThe Ergo-PAS employed synchronized smartphones on the trunk, upper arm, and forearm to capture inertial data during trunk flexion, arm elevation, and repetitive forearm movements. Five participants performed controlled postures, which were validated using a Vicon motion capture system. Pearson's correlation, intraclass correlation coefficients (ICC), and linear regression were applied.ResultsThe Ergo-PAS demonstrated moderate-to-excellent validity and generally good-to-excellent reliability compared with the Vicon motion capture system. Pearson's correlations ranged from 0.13-0.99 for trunk flexion, 0.50-0.95 for upper-arm elevation, and 0.57-0.94 for forearm movements, while ICC values ranged from 0.08-0.99. The performance varied across body segments. Exploratory linear regression calibration showed stable slopes across participants and positions but with joint-specific intercept differences, indicating the need for tailored calibration.ConclusionThis preliminary laboratory validation suggests that the Ergo-PAS can provide reasonably accurate and repeatable posture measurements using smartphone-based inertial sensors. The system shows promise as a low-cost platform for real-time ergonomic risk monitoring and providing feedback. Further research is required to evaluate the usability and effectiveness of these systems during dynamic work activities and in real-world construction environments.
