Deteksi tingkat kelelahan pekerja berbasis smartwatch dengan pendekatan machine learning pada perusahaan manufaktur di Kabupaten Karawang: Mixed method study
DOI: https://doi.org/10.33024/hjk.v18i12.678
Deteksi Kelelahan Pekerja Pendekatan Machine Learning Smartwatch
Abstract
Background: Physical fatigue is one of the most significant and common occupational hazards across various industries. Numerous detection tools, both subjective and objective, have been developed to measure work-related fatigue. Smartwatches are one such tool that can objectively detect physical fatigue by assessing hemodynamic indicators.
Purpose: To detect the level of worker fatigue based on smartwatches with a machine learning approach in manufacturing companies in Karawang Regency.
Method: Mixed explanatory sequential method with fatigue measurement using smartwatch machine learning and Fatigue Severity Scale (FSS).
Results: A total of 42 participants (63%) experienced high levels of fatigue with fatigue scores >42 and an average HR value above 100 times per minute. The Pearson Correlations test produced a correlation coefficient value of 0.039, indicating a strong relationship between HR values and worker fatigue scores. From the results of interviews on worker fatigue levels, 9 themes of worker fatigue experiences were obtained.
Conclusion: Fatigue detection with machine learning that combines HR values and fatigue scores is very effective in preventing work-related fatigue.
Suggestion: Further research is expected to explore the data more deeply. The use of mixed methods can be maintained because it produces varied data, but the addition of the number of participants and research locations needs to be reconsidered. This is to avoid bias levels and produce algorithms from large amounts of data.
Keywords: Detection; Fatigue; Machine Learning Approached; Smartwatch; Workers.
Pendahuluan: Kelelahan fisik adalah salah satu bahaya kerja yang paling penting dan umum terjadi di berbagai industry. Berbagai macam alat telah digunakan untuk mendeteksi/mengukur kelelahan akibat kerja, baik deteksi/pengukuran secara subjektif maupun objektif. Smartwatch adalah salah satu alat yang dapat digunakan untuk mendeteksi tingkat kelelahan fisik secara objektif dengan melihat dari hemodinamik.
Tujuan: Untuk mendeteksi tingkat kelelahan pekerja berbasis smartwatch dengan pendekatan machine learning pada perusahaan manufaktur di Kabupaten Karawang.
Metode: Mixed method explanatory sequential dengan pengukuran kelelahan menggunakan machine learning smartwatch dan Fatigue Severity Scale (FSS).
Hasil: Sebanyak 42 partisipan (63%) mengalami kelelahan tingkat tinggi dengan hasil skor kelelahan >42 dan rata- rata nilai HR diatas 100 kali per menit. Uji Pearson Correlations menghasilkan nilai correlation coefficient sebesar 0.039, menunjukkan ada hubungan kuat antara Nilai HR dengan skor kelelahan pekerja. Dari hasil wawancara tingkat kelelahan pekerja menghasilkan 9 tema pengalaman kelelahan pekerja.
Simpulan: Deteksi kelelahan dengan machine learning yang mengkombinasikan nilai HR dan skor kelelahan sangat efektif dilakukan untuk mencegah terjadinya kelelahan akibat kerja.
Saran: Penelitian selanjutnya diharapkan dapat melakukan eksplorasi data. Penggunaan mixed method dapat dipertahankan karena menghasilkan data yang bervariasi, namun penambahan jumlah partisipan dan lokasi penelitian harus dipertimbangkan ulang. Hal ini untuk menghindari tingkat bias dan menghasilkan algoritma dari jumlah data yang besar.
Kata Kunci: Deteksi; Kelelahan; Pekerja; Pendekatan Machine Learning; Smartwatch.
Downloads
References
AlNoaimi, F. A., & Mazzuchi, T. A. (2021). Risk management application in an oil and gas company for projects. International Journal of Business Ethics and Governance, 1-30.
Baggiano, A., Italiano, G., Guglielmo, M., Fusini, L., Guaricci, A. I., Maragna, R., & Pontone, G. (2022). Changing paradigms in the diagnosis of ischemic heart disease by multimodality imaging. Journal of Clinical Medicine, 11(3), 477.
Belza, B., Miyawaki, C. E., Liu, M., Aree-Ue, S., Fessel, M., Minott, K. R., & Zhang, X. (2018). A systematic review of studies using the multidimensional assessment of fatigue scale. Journal of Nursing Measurement, 26(1), 36-74.
D’Souza, E. (2016). Modified fatigue impact scale–5-item version (MFIS-5). Occupational Medicine, 66(3), 256-257.
Hwang, S., & Lee, S. (2017). Wristband-type wearable health devices to measure construction workers' physical demands. Automation in Construction, 83, 330-340.
Kołodziej, M., Tarnowski, P., Sawicki, D. J., Majkowski, A., Rak, R. J., Bala, A., & Pluta, A. (2020). Fatigue detection caused by office work with the use of EOG signal. IEEE Sensors Journal, 20(24), 15213-15223.
Kurniawati, A. T. (2019). Analisis Faktor yang Mengakibatkan Stres Kerja dan Dampaknya pada Kelelahan Kerja pada Pekerja Divisi Pemeliharaan dan Perbaikan PT. PAL Indonesia (PERSERO) (Doctoral dissertation, Universitas Airlangga).
Mir, M., Nasirzadeh, F., Bereznicki, H., Enticott, P., Lee, S., & Mills, A. (2023). Construction noise effects on human health: Evidence from physiological measures. Sustainable Cities and Society, 91, 104470.
Muttaqin, A. Z., Nawawi, I., & Pamungkas, J. (2024). Implementasi Machine Learning pada Deteksi Kecacatan Printed Circuit Board. Voteteknika (Vocational Teknik Elektronika dan Informatika), 12(1), 44-51.
Rugulies, R., Sørensen, K., Di Tecco, C., Bonafede, M., Rondinone, B. M., Ahn, S., & Pega, F. (2021). The effect of exposure to long working hours on depression: A systematic review and meta-analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury. Environment international, 155, 106629.
Saganowski, S., Perz, B., Polak, A. G., & Kazienko, P. (2022). Emotion recognition for everyday life using physiological signals from wearables: A systematic literature review. IEEE Transactions on Affective Computing, 14(3), 1876-1897.
Seong, S., Park, S., Ahn, Y. H., & Kim, H. (2022). Development of an integrated fatigue measurement system for construction workers: a feasibility study. BMC public health, 22(1), 1593.
Stormorken, E., Jason, L. A., & Kirkevold, M. (2015). Fatigue in adults with post-infectious fatigue syndrome: a qualitative content analysis. BMC nursing, 14, 1-12.
Sudiono, S., Mulyono, S., & Permatasari, H. (2023). Analysis Of Workload And Work-Result Levels On Work Fatigue In Elementary School Teachers In West Karawang District. Jurnal Ners, 7(1), 556-563
Uli, R., & Modjo, R. (2018). Factors Related to Fatigue among Nurses in Jambi Mental Hospital: A 2017 Study. KnE Life Sciences, 190-199.
Portugal, I., Alencar, P., & Cowan, D. (2018). The use of machine learning algorithms in recommender systems: A systematic review. Expert Systems with Applications, 97, 205-227.
Wang, D., & Liu, Q. (2016). Learning to draw samples: With application to amortized mle for generative adversarial learning. arXiv preprint arXiv:1611.01722.
Wong, D. P. L., Chung, J. W. Y., Chan, A. P. C., Wong, F. K. W., & Yi, W. (2014). Comparing the physiological and perceptual responses of construction workers (bar benders and bar fixers) in a hot environment. Applied ergonomics, 45(6), 1705-1711.
Xing, X., Zhong, B., Luo, H., Rose, T., Li, J., & Antwi-Afari, M. F. (2020). Effects of physical fatigue on the induction of mental fatigue of construction workers: A pilot study based on a neurophysiological approach. Automation in Construction, 120, 103381.
Zhang, L., Diraneyya, M. M., Ryu, J., Haas, C. T., & Abdel-Rahman, E. M. (2019). Jerk as an indicator of physical exertion and fatigue. Automation in Construction, 104, 120-128.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Holistik Jurnal Kesehatan
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
