CONCEPTUAL DYNAMIC MODEL OF OPERATOR FUNCTIONAL STATE CONSIDERING THE INTERACTION OF PHYSIOLOGICAL, COGNITIVE, AND OPERATIONAL FACTORS WITH ML-BASED CORRECTION
https://doi.org/10.33815/2313-4763.2026.1.32.100-113
Abstract
This article proposes a conceptual hybrid dynamic model of the operator functional state in maritime transport systems, taking into account the interaction of physiological, cognitive, and operational factors. The relevance of the study is driven by the significant impact of the human factor on maritime safety, particularly the influence of fatigue accumulation and cognitive workload fluctuations on decision-making processes and the occurrence of accidents in complex and time-constrained navigation conditions. Despite extensive research in this domain, most existing approaches to operator state assessment consider these factors separately, and do not provide an integrated and dynamically evolving representation of their interaction, which significantly limits their applicability in real-world operational environments. The proposed model introduces a unified framework for describing the temporal evolution of the operator’s functional state as a multidimensional construct that combines physiological fatigue, cognitive workload, and operational context variables. The model reflects the cumulative and nonlinear nature of fatigue development, the variability and sensitivity of cognitive load to task demands, and the influence of external factors such as traffic density, environmental conditions, and operational complexity. Particular attention is given to the dynamic coupling between these components, which enables capturing feedback effects and interdependencies that are critical for realistic modeling of human performance in maritime systems. A key feature of the approach is the integration of a machine learning-based correction mechanism into the analytical core of the model. Specifically, gradient boosting is employed to approximate residual errors of the base conceptual model, allowing the identification of complex nonlinear relationships and latent dependencies that cannot be explicitly formalized. This hybrid structure ensures a balance between interpretability and predictive capability, preserving the transparency of the model while enhancing its adaptability to empirical data. Although the model is formulated at a conceptual level, it provides a basis for further formalization and numerical implementation in simulation and decision-support environments. The proposed approach can be used as a foundation for intelligent decision support systems in maritime transport, contributing to improved situational awareness, adaptive workload management, and enhanced navigation safety through a more comprehensive and flexible evaluation of operator performance.
References
2. F. Crestelo Moreno, V. Soto-López, J. A. García Maza, M. (2026). Sernaglia, Fatigue as a latent risk factor in maritime safety systems: A systematic review and implications for reliability analysis, Reliability Engineering & System Safety, Volume 267, Part B, 111930, ISSN 0951-8320, https://doi.org/10.1016/j.ress.2025.111930.
3. Matsangas, P., & Shattuck, N. L. (2018). Discriminating Between Fatigue and Sleepiness in the Naval Operational Environment. Behavioral Sleep Medicine, 16(5), 427–436. https://doi.org/10.1080/15402002.2016.1228645.
4. Kerkamm, F., Dengler, D. et al. (2021). Measurement Methods of Fatigue, Sleepiness and Sleep Behaviour Aboard Ships: A Systematic Review. Int J Environ Res Public Health. Vol.19(1):120. https://doi.org/10.3390/ijerph19010120.
5. Ma, M., Liao, R. (2025). Factors affecting seafarers' fatigue: a scoping review. Front Public Health. Vol.13:1647685. https://doi.org/10.3389/fpubh.2025.1647685.
6. Yisi Liu, Zirui Lan, Jian Cui, Gopala Krishnan, Olga Sourina, Dimitrios Konovessis, Hock Eng Ang, Wolfgang Mueller-Wittig (2020). Psychophysiological evaluation of seafarers to improve training in maritime virtual simulator, Advanced Engineering Informatics, Vol.44, 101048, ISSN 1474-0346, https://doi.org/10.1016/j.aei.2020.101048.15 Mao S. et al. (2016).
7. Mao, S., Tu, E, Zhang, G. et al. (2016). An Automatic Identification System (AIS) Database for Maritime Trajectory Prediction and Data Mining, Computer Science > Databases, 1607.03306, https://doi.org/10.48550/arXiv.1607.03306.
8. Moreno, F., Soto-López, V. et al. (2024). Fatigue as a key human factor in complex sociotechnical systems: Vessel Traffic Services. Front. Public Health, Vol.11:1160971. https://doi.org/10.3389/fpubh.2023.116097116.
9. Cao-Feijóo, G., Pérez-Canosa, J. M. et al. (2024). Deep Learning Methods to Mitigate Human-Factor-Related Accidents in Maritime Transport. J. Mar. Sci. Eng., 12, 1819. https://doi.org/10.3390/jmse12101819.
10. Miklody, D., Uitterhoeve, W. M., van Heel, D., Klinkenberg, K., Blankertz, B. (2017). Maritime Cognitive Workload Assessment. Symbiotic Interaction. Symbiotic 2016. Lecture Notes in Computer Science, vol 9961. Springer, Cham. https://doi.org/10.1007/978-3-319-57753-1_9.
11. Main, L. C., Wolkow, A., Chambers, T. P. (2017). Quantifying the Physiological Stress Response to Simulated Maritime Pilotage Tasks: The Influence of Task Complexity and Pilot Experience. J Occup Environ Med. Nov;59(11):1078-1083. https://doi.org/10.1097/JOM. 0000000000001161.
12. NASA/TM—2018–219934. (2018). San Francisco Bar Pilot Fatigue Study, NASA Langley Research Center, Hampton, VA 23681-2199, p. 137 https://ntrs.nasa.gov/api/citations/ 20190002704/ downloads/20190002704.pdf.
13. Tu, E, Zhang, G. et al. (2016). Exploiting AIS Data for Intelligent Maritime Navigation: A Comprehensive Survey, Computer Science. Other Computer Science, Vol. 1606.00981, https://doi.org/10.48550/arXiv.1606.00981.
14. Singh, S. K., Heymann, F. (2020). Machine Learning-Assisted Anomaly Detection in Maritime Navigation Using AIS Data. Electrical Engineering and Systems Science. Signal Processing, Vol. 2002.05013, https://doi.org/10.48550/arXiv.2002.05013.
15. Nguyen, D. et al. (2018). A Multi-task Deep Learning Architecture for Maritime Surveillance using AIS Data Streams. Computer Science. Machine Learning, Vol. 1806.03972,
https://doi.org/10.1109/DSAA.2018.00044.
16. Xue, J., Yang, P. et al. (2025). Machine Learning in Maritime Safety for Autonomous Shipping: A Bibliometric Review and Future Trends. J. Mar. Sci. Eng., 13, 746. https://doi.org/10.3390/jmse13040746.
17. Jiang, Y., Tang, Z., Liu, H. et al. (2026). Work-Life Conditions as the Primary Determinant of Seafarer Mental Health: An Explainable Machine Learning Analysis. INQUIRY: The Journal of Health Care Organization, Provision, and Financing. 2026;63. https://doi.org/10.1177/00469580261438708.
18. Ma, X., Liu, Q. et al. (2024). Machine learning-based multimodal fusion recognition of passenger ship seafarers’ workload: A case study of a real navigation experiment, Ocean Engineering, Vol. 300, 117346, ISSN 0029-8018, https://doi.org/10.1016/j.oceaneng.2024.117346.
19. Zhang, X. et al. (2026). Towards operational safety in maritime transportation: a neurophysiological workload measurement using deep learning, Ocean Engineering.
20. Volume 351, Part 2, (2026). 124399, ISSN 0029-8018, https://doi.org/10.1016/j.oceaneng.2026.124399.
21. Petrovskyi, A. (2025). Dynamic model for calculating navigator fatigue level – Predictive Fatigue Index. Scientific Bulletin of the Kherson State Maritime Academy Vol 31, pp. 199–208, https://doi.org/10.33815/2313-4763.2025.2.31.199-208.
