臺灣博碩士論文加值系統

人工物料搬運(Material Manual Handling, MMH)被認為是工作相關之肌肉骨骼疾病(Work-Related Musculoskeletal Disorders, WMSDs)最主要且常見的危害原因之一，其具有高頻率的重複性、潛在不良的工作姿勢以及過大的工作量等特性都顯著的增加WMSDs的風險，在進行MMH作業的過程中，女性搬運重物時相比於男性可能更容易感到疲勞或受傷，因此是否能夠釐清不同性別之間於MMH作業時面臨的風險差異十分的重要。本研究的目的在於建構一個風險評估模型並應用於MMH作業時的風險評估，同時進一步分析性別與重量於MMH作業時的風險差異。本研究的方法是基於OpenPose與快速全身肢體評估法(Rapid Entire Body Assessment, REBA)，建構隨機森林、支援向量機與極限梯度提升的三種預測模型，並透過均方根誤差(RMSE)、平均絕對誤差(MAE)、決定係數(R2)以及平均絕對百分比誤差(MAPE)進行評估。研究發現三種模型均有足夠的預測性能，其中極限梯度提升在四個評估指標中都有最佳的表現，在統計分析方面，不同性別與不同重量在MMH作業時都有顯著不同的風險，女性在重量增加到10kg以上時的風險會顯著的增加，男性則是增加到20kg以上時才會顯著的增加，不同性別在5kg與10kg上表現有所不同，在20kg時有較為相似的表現，可能顯示當重量超過一定程度時不論何種性別都有增加風險的可能性。本研究基於OpenPose與REBA工具建構風險評估預測模型，並驗證其有效性，且能夠應用於預測MMH作業上。

Manual Material Handling (MMH)　is a leading cause of Work-Related Musculoskeletal Disorders (WMSDs), characterized by repetitive motions, poor postures, and heavy loads, significantly increasing WMSDs risks. Women often experience greater fatigue or injury during MMH compared to men, highlighting the need to understand gender-related risk differences.
This study aims to develop a risk assessment model for MMH and analyze gender and load weight-related risk differences using OpenPose and Rapid Entire Body Assessment (REBA). We constructed three predictive models—Random Forest, Support Vector Machine, and Extreme Gradient Boosting—and evaluated them using RMSE, MAE, R2, and MAPE.
Results indicate all models have sufficient predictive performance, with Extreme Gradient Boosting performing best across all metrics. Statistical analysis shows significant risk differences between genders and load weights in MMH. Women's risk increases significantly at load weights above 10kg, while men's risk increases significantly at 20kg. Gender differences are notable at 5kg and 10kg, but similar at 20kg, suggesting increased risk for both genders at higher load weights.
This study confirms the effectiveness of the OpenPose and REBA-based risk assessment model for predicting MMH risks.

摘要...i
Abstract...ii
誌謝...iv
目錄...v
表目錄...vii
圖目錄...ix
第一章 緒論...1
1.1 研究背景...1
1.2 研究目的...2
1.3 研究流程...3
第二章 文獻探討...4
2.1 人工物料搬運相關研究...4
2.2 性別於與工作相關之肌肉骨骼疾病上的風險差異...5
2.3 評估WMSDs風險的工具...6
2.3.1 OWAS...7
2.3.2 RULA...9
2.3.3 REBA...10
2.4 基於電腦視覺的人體姿勢評估方法...16
2.4.1 Kinect...17
2.4.2 OpenPose...18
2.5 機器學習...21
2.5.1 隨機森林...21
2.5.2 支援向量機...22
2.5.3 極限梯度提升XGBoost...23
第三章 研究方法...25
3.1 原始資料蒐集...25
3.1.1 受試者描述...25
3.1.2 實驗動作...26
3.1.3 負荷重量...27
3.2 資料處理...28
3.2.1 角度計算...28
3.2.2 特徵選取...30
3.2.3 資料集...30
3.3 建立模型...30
3.3.1 隨機森林迴歸...31
3.3.2 支援向量迴歸...33
3.3.3 XGBoost迴歸：...35
3.4 模型評估...38
第四章 研究結果...39
4.1 模型評估與驗證...39
4.2 模型預測...40
4.3 統計分析結果...41
4.3.1 分群比較結果...42
4.3.2 個別比較結果...43
第五章 結論...44
5.1 研究結論...44
5.2 研究限制...44
5.3 未來研究方向...45
參考文獻...46
附錄一...54
附錄二...56
Extended Abstract...67

1.工聯湖畔(2020). <【職安健知多D】小湖話你知「如何正確搬運重物」.>。取自https://www.youtube.com/watch?v=EiEy9dDiNTc 讀取於June 30, 2024.
2.杜珮君、陳秋蓉、張振平 (2023)，我國勞工人體計測調查研究2，勞動部勞動及職業安全衛生研究所，新北市。
3.廖有朋 (2022)，基於EEG腦波訊號建構情緒識別之分類評估模型，國立虎尾科技大學工業管理系工業工程與管理碩士班，碩士論文。
4.潘儀聰、游志雲 (2011)，人因工程現場工作評估及改善流程標準化研究，行政院勞工委員會勞工安全衛生研究所，新北市。
5.Abobakr, A., Nahavandi, D., Hossny, M., Iskander, J., Attia, M., Nahavandi, S., & Smets, M. (2019). RGB-D ergonomic assessment system of adopted working postures. Applied Ergonomics, 80, 75-88.
6.Aktepe, A., Yanık, E., & Ersöz, S. (2021). Demand forecasting application with regression and artificial intelligence methods in a construction machinery company. Journal of Intelligent Manufacturing, 32(6), 1587-1604.
7.Ashish, P. (2019). “Bagging — Ensemble meta Algorithm for Reducing variance”, Medium, available at https://medium.com/ml-research-lab/bagging-ensemble-meta-algorithm-for-reducing-variance-c98fffa5489f retrieved June 30, 2024.
8.Ask Doctor Jo. (2013). “Safe Lifting of Heavy Objects - Ask Doctor Jo”, YouTube, available at https://www.youtube.com/watch?v=zDUg7zHYXxE retrieved June 30, 2024.
9.Atlantic Physical Therapy Center. (2015). “Proper Lifting Techniques”, YouTube, available at https://www.youtube.com/watch?v=PTFBQYxAi_U retrieved June 30, 2024.
10.Benjamin, A. (2017). “Ensemble Learning”, DATA SCIENCE BLOG, available at https://data-science-blog.com/blog/2017/12/03/ensemble-learning/ retrieved June 30, 2024.
11.Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5-32.  
12.Çakıt, E., & Karwowski, W. (2024). Soft computing applications in the field of human factors and ergonomics: A review of the past decade of research. Applied Ergonomics, 114, 104132.
13.Cao, Z., Hidalgo, G., Simon, T., Wei, S. E., & Sheikh, Y. (2021). OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(1), 172-186.
14.Cao, Z., Simon, T., Wei, S. E., & Sheikh, Y. (2017, 21-26 July 2017). Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
15.Center For Total Back Care. (2020). “Proper Lifting and Bending Techniques”, YouTube, available at https://www.youtube.com/watch?v=2S5T6Lx7LF4 retrieved June 30, 2024.
16.Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA.
17.Clark, R. A., Mentiplay, B. F., Hough, E., & Pua, Y. H. (2019). Three-dimensional cameras and skeleton pose tracking for physical function assessment: A review of uses, validity, current developments and Kinect alternatives. Gait & Posture, 68, 193-200.
18.Coenen, P., Gouttebarge, V., van der Burght, A. S. A. M., van Dieën, J. H., Frings-Dresen, M. H. W., van der Beek, A. J., & Burdorf, A. (2014). The effect of lifting during work on low back pain: a health impact assessment based on a meta-analysis. Occupational and Environmental Medicine, 71(12), 871.
19.Continuing Care Safety Association. (2023). “MIP Techniques – Lifting Awkward or Uneven Loads”, YouTube, available at https://www.youtube.com/watch?v=eb2R-7R_hGI&list=PLe2v0CS1bT1zHdrbQ6NG4J3hZxtDW_A6n&index=12 retrieved June 30, 2024.
20.Corporate Work Health Australia. (2020). “Manual handling: Twisting and turning”, YouTube, available at https://www.youtube.com/watch?v=Qjqqds6QIac retrieved June 30, 2024. 
21.da Costa, B. R., & Vieira, E. R. (2010). Risk factors for work-related musculoskeletal disorders: a systematic review of recent longitudinal studies. American Journal of Industrial Medicine, 53(3), 285-323.
22.Das, S., Krishna Moorthy, M., & Shanmugaraja, K. (2023). Analysis of Musculoskeletal Disorder Risk in Cotton Garment Industry Workers. Journal of Natural Fibers, 20(1), 2162182.
23.David, G. C. (2005). Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders. Occup Med (Lond), 55(3), 190-199.
24.Diego-Mas, J. A., & Alcaide-Marzal, J. (2014). Using Kinect™ sensor in observational methods for assessing postures at work. Applied Ergonomics, 45(4), 976-985.
25.Dzeng, R. J., Hsueh, H. H., & Ho, C. W. (2017, 22-26 May 2017). Automated Posture Assessment for construction workers. 2017 40th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO).
26.Elite Force Safety. (2022). “Avoiding Injury: Understanding the Dangers of Twisting when Lifting - Manual Handling Tips”, YouTube, available at https://www.youtube.com/watch?v=b06K4u_Er_g retrieved June 30, 2024.
27.Fox, R. R., Lu, M.-L., Occhipinti, E., & Jaeger, M. (2019). Understanding outcome metrics of the revised NIOSH lifting equation. Applied Ergonomics, 81, 102897.
28.Gómez-Galán, M., Pérez-Alonso, J., Callejón-Ferre Á, J., & López-Martínez, J. (2017). Musculoskeletal disorders: OWAS review. Ind Health, 55(4), 314-337.
29.Heran-Le Roy, O., Niedhammer, I., Sandret, N., & Leclerc, A. (1999). Manual materials handling and related occupational hazards: a national survey in France. International Journal of Industrial Ergonomics, 24(4), 365-377.
30.Hignett, S., & McAtamney, L. (2000). Rapid Entire Body Assessment (REBA). Applied Ergonomics, 31(2), 201-205.
31.Hita-Gutiérrez, M., Gómez-Galán, M., Díaz-Pérez, M., & Callejón-Ferre Á, J. (2020). An Overview of REBA Method Applications in the World. Int J Environ Res Public Health, 17(8).
32.Hoozemans, M. J. M., Kingma, I., de Vries, W. H. K., & van Dieën, J. H. (2008). Effect of lifting height and load mass on low back loading. Ergonomics, 51(7), 1053-1063.
33.Huang, K., Jia, G., Wang, Q., Cai, Y., Zhong, Z., & Jiao, Z. (2024). Spatial relationship-aware rapid entire body fuzzy assessment method for prevention of work-related musculoskeletal disorders. Applied Ergonomics, 115, 104176.
34.Jiang, F., Lu, Y., Chen, Y., Cai, D., & Li, G. (2020). Image recognition of four rice leaf diseases based on deep learning and support vector machine. Computers and Electronics in Agriculture, 179, 105824.
35.Joshi, M., & Deshpande, V. (2020). Investigative study and sensitivity analysis of Rapid Entire Body Assessment (REBA). International Journal of Industrial Ergonomics, 79, 103004.
36.Karhu, O., Kansi, P., & Kuorinka, I. (1977). Correcting working postures in industry: A practical method for analysis. Applied Ergonomics, 8(4), 199-201.
37.Kim, W., Sung, J., Saakes, D., Huang, C., & Xiong, S. (2021). Ergonomic postural assessment using a new open-source human pose estimation technology (OpenPose). International Journal of Industrial Ergonomics, 84, 103164.
38.Ko, D., & Park, S. (2024). Investigating the Correlation between Air Pollution and Housing Prices in Seoul, South Korea: Application of Explainable Artificial Intelligence in Random Forest Machine Learning. Sustainability, 16(11), 4453.
39.Lee, Y.-C., & Lee, C.-H. (2022). SEE: A proactive strategy-centric and deep learning-based ergonomic risk assessment system for risky posture recognition. Advanced Engineering Informatics, 53, 101717.
40.Li, X., & Matthies, D. J. C. (2023). ShoeTect: Detecting Body Posture, Ambulation Activity, Gait Abnormalities, and Terrain with Multisensory Smart Footwear Proceedings of the 7th International Workshop on Sensor-based Activity Recognition and Artificial Intelligence, Rostock, Germany.
41.Li, Y.-S., Pai, P.-F., & Lin, Y.-L. (2023). Forecasting inflation rates be extreme gradient boosting with the genetic algorithm. Journal of Ambient Intelligence and Humanized Computing, 14(3), 2211-2220.  
42.Lin, P.-C., Chen, Y.-J., Chen, W.-S., & Lee, Y.-J. (2022). Automatic real-time occupational posture evaluation and select corresponding ergonomic assessments. Scientific Reports, 12(1), 2139.
43.Lunin, A., & Glock, C. H. (2021). Systematic review of Kinect-based solutions for physical risk assessment in manual materials handling in industrial and laboratory environments. Computers & Industrial Engineering, 162, 107660.
44.Manghisi, V. M., Uva, A. E., Fiorentino, M., Bevilacqua, V., Trotta, G. F., & Monno, G. (2017). Real time RULA assessment using Kinect v2 sensor. Applied Ergonomics, 65, 481-491.
45.Mazhar, O., Navarro, B., Ramdani, S., Passama, R., & Cherubini, A. (2019). A real-time human-robot interaction framework with robust background invariant hand gesture detection. Robotics and Computer-Integrated Manufacturing, 60, 34-48.
46.Nakano, N., Sakura, T., Ueda, K., Omura, L., Kimura, A., Iino, Y., Fukashiro, S., & Yoshioka, S. (2020). Evaluation of 3D Markerless Motion Capture Accuracy Using OpenPose With Multiple Video Cameras. Front Sports Act Living, 2, 50.
47.Nayak, G. K., & Kim, E. (2021). Development of a fully automated RULA assessment system based on computer vision. International Journal of Industrial Ergonomics, 86, 103218.
48.Ni, S., Peng, Y., Peng, K., & Liu, Z. (2022). Supply Chain Demand Forecast Based on SSA-XGBoost Model. Journal of Computer and Communications, 10, 71-83.
49.Nimbarte, A. D. (2014). Risk of neck musculoskeletal disorders among males and females in lifting exertions. International Journal of Industrial Ergonomics, 44(2), 253-259.
50.Nogueira, H. C., Locks, F., Barbieri, D. F., & Oliveira, A. B. (2018). How does the biomechanical exposure of the upper body in manual box handling differ from exposure in other tasks in the real industrial context? International Journal of Industrial Ergonomics, 68, 8-14.  
51.Ota, M., Tateuchi, H., Hashiguchi, T., & Ichihashi, N. (2021). Verification of validity of gait analysis systems during treadmill walking and running using human pose tracking algorithm. Gait & Posture, 85, 290-297.
52.Özen, F. (2024). Random forest regression for prediction of Covid-19 daily cases and deaths in Turkey. Heliyon, 10(4), e25746.
53.Plantard, P., Shum, H. P. H., Le Pierres, A.-S., & Multon, F. (2017). Validation of an ergonomic assessment method using Kinect data in real workplace conditions. Applied Ergonomics, 65, 562-569.
54.PhoenixParks. (2012). “Safe Lifting”, YouTube, available at https://www.youtube.com/watch?v=901uQgfiuVk retrieved June 30, 2024.
55.Punitha, N., Divya Bharathi, K., Manuskandan, S. R., & Karthick, P. A. (2024). Analysis of Muscle Fatigue Progression Using Geometric Features of Surface Electromyography Signals and Explainable XGBoost Classifier. Journal of Medical and Biological Engineering, 44(2), 191-197.
56.Ribeiro, N. F., & Santos, C. P. (2017, 16-18 Feb. 2017). Inertial measurement units: A brief state of the art on gait analysis. 2017 IEEE 5th Portuguese Meeting on Bioengineering (ENBENG).
57.Safety First Training. (2014). “Lifting boxes at bench level by Safety First Training.”, YouTube, available at https://www.youtube.com/watch?v=NuutewGhwxM retrieved June 30, 2024.
58.Safety First Training. (2014). “Lifting boxes from the floor by Safety First Training.”, YouTube, available at https://www.youtube.com/watch?v=1rHQLzdYvL4 retrieved June 30, 2024.
59.SAFEWorkManitoba. (2021). “Safe Lifting in the Workplace”, YouTube, available at https://www.youtube.com/watch?v=251OvQNaVVg retrieved June 30, 2024.
60.Santa Fe College Teaching Zoo. (2022). “50 Pound Bag Lift Demonstration - Zoo Animal Technology program”, YouTube, available at https://www.youtube.com/watch?v=TfZbbx0zyg0 retrieved June 30, 2024.
61.Schneider, E., Irastorza, X., Safety, E., Copsey, S., Verjans, H., Eeckelaert, L., Safety, V., & Belgium. (2010). OSH in Figures: Work-related Musculoskeletal Disorders in the EU — Facts and Figures.
62.Schwartz, A. H., Albin, T. J., & Gerberich, S. G. (2019). Intra-rater and inter-rater reliability of the rapid entire body assessment (REBA) tool. International Journal of Industrial Ergonomics, 71, 111-116.
63.Skals, S., Bláfoss, R., de Zee, M., Andersen, L. L., & Andersen, M. S. (2021). Effects of load mass and position on the dynamic loading of the knees, shoulders and lumbar spine during lifting: a musculoskeletal modelling approach. Applied Ergonomics, 96, 103491.
64.Srinivasan, D., Sinden, K. E., Mathiassen, S. E., & Côté, J. N. (2016). Gender differences in fatigability and muscle activity responses to a short-cycle repetitive task. Eur J Appl Physiol, 116(11-12), 2357-2365.
65.Stenum, J., Rossi, C., & Roemmich, R. T. (2021). Two-dimensional video-based analysis of human gait using pose estimation. PLOS Computational Biology, 17(4), e1008935.
66.Swamynathan, M. (2017). Mastering Machine Learning with Python in Six Steps.
67.Tran, T.-a., Ruppert, T., Eigner, G., & Abonyi, J. (2023). Assessing human worker performance by pattern mining of Kinect sensor skeleton data. Journal of Manufacturing Systems, 70, 538-556.
68.University of California | Risk & Safety Training. (2018). “Safe Lifting: Low Lift & Transfer 09/25/18”, YouTube, available at https://www.youtube.com/watch?v=r1EspSH1EbI retrieved June 30, 2024.
69.U-Pack. (2015). “How to Lift Boxes..”, YouTube, available at https://www.youtube.com/watch?v=2cvAn1kzVLU retrieved June 30, 2024.
70.Vadivel, Subramaniyan, C., Muthukumar, K., & Bharani, T. (2022). Ergonomics hazards and risk assessment for the saw mill operations. Materials Today: Proceedings, 66, 717-724.
71.Vignais, N., Miezal, M., Bleser, G., Mura, K., Gorecky, D., & Marin, F. (2013). Innovative system for real-time ergonomic feedback in industrial manufacturing. Applied Ergonomics, 44(4), 566-574.
72.Vishwakarma, D. K., & Singh, T. (2019). A visual cognizance based multi-resolution descriptor for human action recognition using key pose. AEU - International Journal of Electronics and Communications, 107, 157-169.
73.Washabaugh, E. P., Shanmugam, T. A., Ranganathan, R., & Krishnan, C. (2022). Comparing the accuracy of open-source pose estimation methods for measuring gait kinematics. Gait & Posture, 97, 188-195.
74.Widanarko, B., Legg, S., Stevenson, M., Devereux, J., Eng, A., Mannetje, A. t., Cheng, S., Douwes, J., Ellison-Loschmann, L., McLean, D., & Pearce, N. (2011). Prevalence of musculoskeletal symptoms in relation to gender, age, and occupational/industrial group. International Journal of Industrial Ergonomics, 41(5), 561-572.
75.Xu, Q., Huang, G., Yu, M., & Guo, Y. (2020). Fall prediction based on key points of human bones. Physica A: Statistical Mechanics and its Applications, 540, 123205.
76.Yang, F., Di, N., Guo, W.-w., Ding, W.-b., Jia, N., Zhang, H., Li, D., Wang, D., Wang, R., Zhang, D., Liu, Y., Shen, B., Wang, Z.-x., & Yin, Y. (2023). The prevalence and risk factors of work related musculoskeletal disorders among electronics manufacturing workers: a cross-sectional analytical study in China. BMC Public Health, 23(1), 10.
77.Yang, S. T., Park, M. H., & Jeong, B. Y. (2020). Types of manual materials handling (MMH) and occupational incidents and musculoskeletal disorders (MSDs) in motor vehicle parts manufacturing (MVPM) industry. International Journal of Industrial Ergonomics, 77, 102954.
78.Yu, N., Hong, L., & Guo, J. (2021). Analysis of upper-limb muscle fatigue in the process of rotary handling. International Journal of Industrial Ergonomics, 83, 103109.
79.Zadeh, L. A. (1994). Soft computing and fuzzy logic. IEEE Software, 11(6), 48-56.