Machine-Learning-Based Fatigue Trend Analysis on IMU Wearable Sensor Data from Construction Site Workers - Sensors 2025

Aug 5, 2026

Upload by Aurelia

Physical fatigue drives construction-site accidents and musculoskeletal injuries, yet practical tools for detection and management remain limited. Wearable inertial measurement units (IMUs) can detect fatigue, but prior validation has largely stayed in laboratory environments. This study uses IMU data from real site workers with simulated tasks to support fatigue trend detection in real-life conditions, adds frequency-domain feature analysis, and applies machine learning to improve accuracy while identifying effective sensor locations and features.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Motivation and research gap

Existing fatigue detection approaches

Role of IMU sensors and privacy considerations

Inspiration from prior work

Article Machine-Learning-Based Fatigue Trend Analysis on IMU Wearable Sensor Data from Construction Site Workers Janne S. Keränen 1, *, Jamil Ahmad 2,3, Sergio Leggieri 2, Satu-Marja Mäkelä 1, Darwin G. Caldwell 2, Christian Di Natali 2, Atte Kinnula 1 and Pekka Siirtola 4 Academic Editor: Kenneth Loh Received: 29 September 2025 Revised: 21 November 2025 Accepted: 2 December 2025 Published: 8 December 2025 Citation: Keränen, J.S.; Ahmad, J.; Leggieri, S.; Mäkelä, S.-M.; Caldwell, D.G.; Di Natali, C.; Kinnula, A.; Siirtola, P. Machine-Learning-Based Fatigue Trend Analysis on IMU Wearable Sensor Data from Construction Site Workers. Sensors 2025, 25, 7455. [https://doi.org/](https://doi.org/)[ ](https://doi.org/)[10.3390/s25247455](10.3390/s25247455) Copyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://creativecommons.org/](https://creativecommons.org/)[ ](https://creativecommons.org/)[licenses/by/4.0/](licenses/by/4.0/)) . 1 VTT Technical Research Centre of Finland Ltd., Kaitoväylä 1, P.O. Box 1100, 90571 Oulu, Finland; satu-marja.makela@vtt.fi (S.-M.M.); atte.kinnula@vtt.fi (A.K.) 2 Department of Advanced Robotics (ADVR), Istituto Italiano di Tecnologia, via Morego, 30, 16163 Genoa, Italy; jamil.ahmad@iit.it (J.A.); sergio.leggieri@iit.it (S.L.); darwin.caldwell@iit.it (D.G.C.); christian.dinatali@iit.it (C.D.N.) 3 Department of Informatics, Bioengineering, Robotics and Systems Engineering (DIBRIS), Università degli Studi di Genova (UniGe), 16145 Genova, Italy 4 Biomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Pentti Kaiteran katu 1, P.O. Box 8000, 90014 Oulu, Finland; [pekka.siirtola@oulu.fi](pekka.siirtola@oulu.fi) * [Correspondence: janne.s.keranen@vtt.fi](Correspondence: janne.s.keranen@vtt.fi) Abstract Physical fatigue is a major cause of work-related accidents and musculoskeletal injuries in the construction industry, and additional means are needed for their identification and management to prevent long-term consequences. Based on recent scientific literature, fatigue can be detected with wearable inertial measurement units (IMUs) . However, IMUs for detecting fatigue have been so far tested mainly in the laboratory; therefore, a research gap exists in application of IMU sensors for detecting fatigue in real-life work settings. The aim of this paper is to bring the fatigue trend detection with IMUs closer to real-life context by using wearable IMU sensor data from an actual construction site measuring actual workers with simulated work tasks. The paper also presents advancements in fatigue trend detection with frequency domain investigations to gain access to more detailed fatigue relevant features. Machine-learning methods are used to predict fatigue trends based on IMU data, resulting in fatigue trend detection accuracy that advances the state of the art. More knowledge is also unearthed about relevant sensor locations and features. Keywords: fatigue; wearables; sensor; IMU; machine learning 1. Introduction In the US construction industry, 33% of all work-related musculoskeletal injuries are attributed to overexertion [1] . In addition, physical fatigue is found to be a major cause of work-related accidents in the building construction industry [2,3] . Therefore, detecting and tracking fatigue is important, so that timely interventions (e.g., breaks) can be introduced [4] . To support physical ergonomics and overall safety at work, personalized fatigue assessments would be beneficial. Fatigue is usually detected by measuring heart rate (HR) and heart rate variability (HRV), skin temperature, muscle jerk (IMU) and muscle response (electromyography, EMG) metrics [5] . However, HR, HRV, skin temperature and EMG all require a sensor with skin conta

FAQs

Why is detecting physical fatigue important in construction work?

Physical fatigue is linked to work-related accidents and musculoskeletal injuries, so timely interventions such as breaks can reduce long-term consequences.

What gap does this paper address regarding IMU-based fatigue detection?

Most IMU fatigue detection research has been tested mainly in laboratory settings, leaving limited evidence for real-life construction environments.

How does the study improve fatigue trend detection using IMU data?

It applies machine-learning methods for fatigue trend prediction and extends analysis with frequency-domain investigations to extract more detailed fatigue-relevant features, also examining sensor locations and features.

Aug 5, 2026

Upload by Aurelia