Wu, Lun-WangLun-WangWuChen, Kun-YiKun-YiChenHUI-PING TSERNG2025-08-282025-08-282025-08-1200140139https://www.scopus.com/record/display.uri?eid=2-s2.0-105012972205&origin=resultslisthttps://scholars.lib.ntu.edu.tw/handle/123456789/731672This study conducted a non-interventional longitudinal study to collect physiological data from tunnel construction workers using an Internet-of-Things-based photoplethysmography physiological monitoring system. The analysis results indicated that among tunnel workers, the SDNN and RMSSD of muck car drivers were significantly lower than those of other occupations, reflecting decreased sympathetic innervation or parasympathetic activity, which is typical of mental stress. During tunnel construction, many HRV indicators of workers were significantly higher in the morning and afternoon than during the lunch break. The most obvious physiological response to noise was observed in the muck car driver, where the SDNN, LF, and LF/HF indices increased significantly under noise >85dB. This study fills an empirical gap in the objective assessment of mental stress among tunnel workers, providing a reference for optimising work scheduling and task allocation. It provides a framework for occupational health monitoring and stress management strategies.falseheart rate variability (HRV)Internet-of-Things (IoT)mental stressTunnel construction[SDGs]SDG3journal article10.1080/00140139.2025.25435182-s2.0-105012972205