Evaluating a Wearable Sensor System for Musculoskeletal Risk Assessment in Simulated Manual Assembly Line Tasks

Read full text Summarise with VIRAT
Authors: Daniel Bal
Cited by: 0
References: 52

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Wearable inertial measurement units (IMUs) promise low-cost, field-ready musculoskeletal risk assessment, but their validity against laboratory gold standards remains task-dependent and requires systematic evaluation alongside deployment readiness. The aim of this study was to evaluate the validity of a commercial IMU system (Wergonic Smart Tshirt) for measuring upper arm elevation and trunk flexion against optical motion capture (Vicon), and to assess system usability for occupational deployment. A mixed-methods design was employed, where kinematic validation (n = 1) compared Wergonic and Vicon measurements across five simulated assembly-line tasks. Due to sensor placement constraints preventing simultaneous recording, data were collected in separate, protocol-matched sessions. Agreement was assessed distributionally (percentage of time above posture thresholds) and by activity-pattern timing. Usability evaluation (n = 6) employed the Post-Study System Usability Questionnaire (PSSUQ), supplemented by semi-structured interviews (n = 2) analysed using the Framework Method. Given the small sample sizes and non-simultaneous recording sessions, this study should be considered a proof-of-concept pilot requiring confirmation through larger-scale field validation. The results showed that in upper arm elevation measurement, Wergonic reported approximately 2.4 times lower exposure percentages above 90° than Vicon. Despite this quantitative difference, both systems produced identical risk classifications (κ = 1.00) when applying a threshold-based criteria. For trunk flexion, large task-dependent biases (ranging from 15° to 62°) compromised quantitative agreement; however, activity timing alignment remained acceptable (81–87% peak agreement). Usability assessment revealed an overall PSSUQ score of 3.54, with the three subscales showing a hierarchical pattern: Interface Quality (assessing screen organisation and pleasantness; M = 3.00) outperformed Information Quality (assessing clarity and effectiveness of information; M = 3.58) and System Usefulness (assessing ease and efficiency of task completion; M = 3.75). Interviews identified timestamp accuracy, session naming, and repeated calibration requirements as key deployment barriers. In conclusion, the Wergonic system demonstrates acceptable validity for binary upper arm elevation risk classification, supporting its use for screening overhead work exposure. However, trunk flexion measurements showed systematic biases too large for quantitative exposure assessment. The identified usability barriers represent addressable engineering issues rather than fundamental sensor limitations. For threshold-based risk decisions, the system delivers appropriate judgements; for precise angle quantification, current implementation lacks sufficient reliability.

Fields of Study & Subjects

Ergonomics and Musculoskeletal Disorders Context-Aware Activity Recognition Systems Occupational Health and Safety in Workplaces

Publication details

Journal / source
Diva portal (Dalarna University Library)
Publication date
1 January 2026
Language
EN
Document type
Thesis