以PPG心率手環及心率變異數分析評估隧道工程勞工精神負荷

Application of PPG Wristband on Fatigue and Stress Evaluation of Tunnel Construction Using Heart Rate Variability

陳錕鎰(Kun-Yi Chen)

指導教授 ： 曾惠斌

國立臺灣大學/工學院/土木工程學研究所/碩士(2020年)

https://doi.org/10.6342/NTU202100155

摘要

營造業現場勞工安全事故一直是土木工程產業的重要議題，尤其高風險的特殊環境施工作業所造成的職業災害更是不容小覷。根據勞動部重大職業災害2009-2018年統計數據顯示，營建產業僱傭人數只佔整體產業的9%，卻造成整體職業災害中致死人數的47%。並根據其細部統計，近年來營造業現場中之侷限空間造成的職災比例逐年升高，其中隧道施工場所亦為此類現場，因此屬高風險場域。造成勞工安全事故發生的原因，其中以疲勞與精神壓力最為普遍，在長時間精神壓力下亦會累積疲勞程度。因此，加強勞工精神壓力的管理，預先在可能產生意外的情況下警示勞工，防患未然，有效降低勞安事故的發生機率，是本篇論文所要探討之重點。根據文獻得知，精神壓力可藉由個人心理指標心率變異數(Heart Rate Varability，HRV)計算，而心率則是運算心率變異數指標中數據的起點。因此，本研究引進目前新興科技：光體積變化描記圖法(Photoplethysmography，PPG) 技術之心率感測手環，在不影響勞工施作工項的條件下，於隧道施作現場進行數據取樣，收集現場勞工的心率時序變化資料。本研究使用國科會相關研究「隧道工程局限空間環境勞工生理監測策略及管理系統之研究」中研究人員陳維政博士生收集之數據，以當事人知情同意下進行數據後續處理。數據收集隧道施工中重點8個工項進行一個月數據收集，後續心率變異數各指標以KubiosHRV軟體運算。評估以不同工項、工時、環境條件(噪音)下，勞工的心理壓力程度。依據本研究結果發現，長時間施作隧道內部工項之勞工心理壓力程度較高，因此判斷為重點監測族群；而工作時段以下午時段顯示勞工心理壓力程度較高，推斷因工作負荷的累積與下午工時較長導致；噪音部分，根據統計結果顯示，受噪音影響之工項勞工中，高噪音程度與低程度之心率變異數指標的平均值無明顯差距，但以台車運輸手及潛盾機操作手心率變異數極距在高噪音環境下相對較低，推估可能原因為高噪音發生條件時為這兩項工項暫停作業階段，因此心理狀態變動較小。
本研究因數據收集現場限制造成數據中斷情形及設備本身限制，後續數據以內插方式補上中間缺漏數據，分析過程已盡可能選擇完整數據單元進行運算。因此未來研究若能夠解決數據收集的種種限制條件，並加上原本心率手環的資料儲存空間，將能夠更進一步得到更多筆的資料提供後續分析。

關鍵字

PPG ； 隧道工程 ； 心率變異數 ； Kubios ； 工作負荷

並列摘要

he issue of occupational accidents in construction worksite especially for high-risk activities is always a great concern in the civil engineering industry. According to the statistical yearbook, Minister of Labor, among the number of employees within all the occupation, there are merely 9% working in construction engineering. However, it has caused almost 47% of the number of fatality. Especially in some hazardous occupations such as tunnel construction, the number has accelerated gradually. Among all the factors, fatigue and stress are the commonest. They can be detected in both physiological and psychological ways. In this paper, heart rate variability (HRV) analysis is applied, which has been proved as a representation of the psychological state, utilizing PPG wristbands, medical signal analyze software Kubios and statistical methods. The data set was collected by a senior, Wei Cheng Chen. The usage of the data set has been authorized by him officially. Data collection included 8 work tasks in the construction worksite of a shield tunnel for a month of data collection. Parameters of HRV were measured during the daytime of work from 7a.m to 7p.m. The data then be evaluated and categorized according to different tasks, work period and environmental condition, which is the noise. According to the result, we can identify that tasks which need to be conducted inside the tunnel pile up more psychological stress to the labors. In condition, the value of HRV parameters appear to be higher during the afternoon period than the ones in the morning. We can then speculate that due to the accumulative tiredness accompanied by a longer period of working hours, the labors experience more psychological stress during the afternoon hours. For the noise evaluation, within the labors that were influenced by noise condition, the mean value of their HRV parameters are surprisingly similar. However, we can see the span of them are smaller during the high-noise condition when their works are paused. Thus, their HRV parameters are relatively stable. With these result and observation, we can detect which task and when the labors feel more stressful which might cause accident. Therefore, the construction managing teams are able to prevent it in advance.
In this research, due to the limitation of the site and the uncertainty of the whole crews, the condition of data collection was not as well as we expected. Thus, we were forced to compensate the lacked data by further calculation. However, in the research, most of the data span we analyzed are the one with best completeness. If the future research is able to solve these limitations, more data can be analyzed which means the result will be even more accurate.

並列關鍵字

PPG ； tunnel construction ； HRV ； Kubios ； workload

參考文獻

(2018). 107年勞動檢查年報. 勞動部職業安全衛生署.

Ahamed, T. S., Islam, T. M. (2016). An efficient method for heart rate monitoring using wrist-type photoplethysmographic signals during intensive physical exercise. 2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), (pp. 863-868). Dhaka.

Chao PC, Juang YJ, Chen CJ, Dai YT, Yeh CY, Hu CY. (2013, 10). Combined effects of noise, vibration, and low. Kaohsiung Journal of Medical Sciences(10), pp. 560-567.

Deus LA, Sousa CV, Rosa TS, Filho JMS, Santos PA, Barbosa LD, Silva Aguiar S, Souza LHR, Simões HG. (2019, 6). Heart rate variability in middle-aged sprint and endurance athletes. Physiology Behavior(205), pp. 39-43.

Evrengul H, Tanriverdi H, Kose S, Amasyali B, Kilic A, Celik T, Turhan H. (2006, 4). The relationship between heart rate recovery and heart rate variability in coronary artery disease. Annals of Noninvasive Electrocardiology, 11(2), pp. 154-162.