An evaluation of wearable sensors and their placements for analyzing construction worker's trunk posture in laboratory conditions. (November 2017)

• Record Type: Journal Article
• Authors: Lee, Wonil; Seto, Edmund; Lin, Ken-Yu; Migliaccio, Giovanni C.
• Is Part Of: Applied ergonomics. Volume 65(2017)
• Page Start: 424
• Page End: 436
• Publication Date: 2017-11
• Subjects: Work-related musculoskeletal disorder -- Accelerometer for inclinometry -- Construction worker; Human engineering -- Periodicals; 620.82
• DOI: 10.1016/j.apergo.2017.03.016
• Languages: English
• ISSNs: 0003-6870

Abstract

Abstract: This study investigates the effect of sensor placement on the analysis of trunk posture for construction activities using two off-the-shelf systems. Experiments were performed using a single-parameter monitoring wearable sensor (SPMWS), the ActiGraph GT9X Link, which was worn at six locations on the body, and a multi-parameter monitoring wearable sensor (MPMWS), the Zephyr BioHarness™3, which was worn at two body positions. One healthy male was recruited and conducted 10 experiment sessions to repeat measurements of trunk posture within our study. Measurements of upper-body thoracic bending posture during the lifting and lowering of raised deck materials in a laboratory setting were compared against video-captured observations of posture. The measurements from the two sensors were found to be in agreement during slow-motion symmetric bending activities with a target bending of ≤45°. However, for asymmetric bending tasks, when the SPMWS was placed on the chest, its readings were substantially different from those of the MPMWS worn on the chest or under the armpit. Highlights: We compared the wearable accelerometer placements for trunk posture analysis. We evaluated multi-parameter and single-parameter monitoring wearable sensor (MPMWS and SPMWS) systems. MPMWS worn under the armpit and SPMWS worn on the chest showed comparable results in slow and moderate symmetric trunk flexion. MPMWS worn on the chest and under the armpit were partially comparable to the video-based observation in asymmetric bending. Under the specified conditions, wearable sensors are a promising technology for ergonomics research.