A Physics-Informed Multi-Modal Deep Learning Framework for Stochastic Prediction and Reliability Assessment of Human-Induced Vibrations in Lightweight and Composite Floor Systems

References

[1] Wilk-Jakubowski, J.L., Pawlik, L., Frej, D., Wilk-Jakubowski, G. The evolution of machine learning in vibration and acoustics: A decade of innovation (2015–2024). Applied Sciences 15, 6549 (2025). Crossref

[2] Shahabpoor, E., Berari, B., Pavic, A. Vibration serviceability assessment of floor structures: Simulation of human–structure–environment interactions using agent-based modeling. Sensors 25, 126 (2025). Crossref

[3] Nirgude, V.N., Majid, A., Vasanthi, P., et al. Machine learning on structural vibration time-series signals for structural stability prediction and seismic performance evaluation. International Journal of Advanced Signal and Image Sciences 12(2s), 836–847 (2026).

[4] Singh, D. Application of machine learning in civil engineering: Review. Advancements in Civil Engineering & Technology 6, 000639 (2024). Crossref

[5] Wang, L., Gong, G., Xia, J. Comparison of evaluations of human-induced floor vibration based on different design guidelines. Structures 68, 107114 (2024). Crossref

[6] López Cela, J.J., Martínez Vicente, J.L. Vibration in floors induced for human walking: Comparison of two design guidelines. Buildings 14, 3911 (2024). Crossref

[7] Bakalis, K., Kazantzi, A. Composite floors under human-induced vibrations. In: Proceedings of Structural Engineering Conference (2023).

[8] Pu, X., He, T., Zhu, Q. Human-induced vibration serviceability analysis of high-frequency floors under different layouts of obstacles based on random loading and random structural properties. Computers & Structures 316, 107895 (2025). Crossref

[9] Liang, H., Lu, Y., Xie, W., He, Y., Wei, P., Zhang, Z., Wang, Y. Prediction of human-induced structural vibration using multi-view markerless 3D gait reconstruction and an enhanced bipedal human–structure interaction model. Journal of Sound and Vibration 602, 118931 (2025). Crossref

[10] Li, J., Tang, W., Liu, J., Zhao, Y., Chen, Y.F. EEG-based floor vibration serviceability evaluation using machine learning. Advanced Engineering Informatics 64, 103089 (2025). Crossref

[11] Dong, Y., Noh, H.Y. Continual person identification using footstep-induced floor vibrations on heterogeneous floor structures. In: Proceedings of Structural Health Monitoring Conference (2025). https://arxiv.org/abs/2502.15632

[12] Zhang, Y., Liang, X. Domain adaptation in structural health monitoring of civil infrastructure: A systematic review. arXiv preprint arXiv:2512.18780 (2025).

[13] Chandra, A.R.A., Mahesh, S., Ravikumar, L., Srivatsan, T.S. An overview on application of machine learning to emerging frontiers in civil engineering science. JOJ Material Science 8(5), 555747 (2024). Crossref

[14] Gallet, A. Machine learning for structural design models from the inverse problem perspective. PhD Thesis, University of Sheffield (2024).

[15] Avci, O., Abdeljaber, O., Kiranyaz, S. Transfer learning in structural health monitoring. In: Dynamics of Civil Structures, Volume 2 (Springer, 2021).

[16] Meera, M.A., Sandeep, M.N. Predictive modeling of ground vibration induced by high-speed train using machine learning techniques. In: Proceedings of CEASIDE 2025. Crossref

[17] Wang, L., Gong, G., Xia, J. Comparison of evaluations of human-induced floor vibration based on different design guidelines. Structures 68, 107114 (2024). Crossref

[18] López Cela, J.J., Martínez Vicente, J.L. Vibration in floors induced for human walking: Comparison of two design guidelines. Buildings 14, 3911 (2024). Crossref

[19] Pu, X., He, T., Zhu, Q. Human-induced vibration serviceability analysis. Computers & Structures 316, 107895 (2025). Crossref

[20] Liang, H., et al. Prediction of human-induced structural vibration. Journal of Sound and Vibration 602, 118931 (2025). Crossref

[21] https://www.kaggle.com/datasets/ziya07/building-structural-health-sensor-dataset

@article{1722945,
      author = {Ernest Chidindu Ernest, Dr. Chen Deyi},
      title = {A Physics-Informed Multi-Modal Deep Learning Framework for Stochastic Prediction and Reliability Assessment of Human-Induced Vibrations in Lightweight and Composite Floor Systems},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {10},
      number = {3},
      pages = {1287-1315},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1722945.pdf},
      abstract = {Human-induced vibrations in lightweight and composite floor systems pose important serviceability concerns because these structures typically have lower stiffness, reduced mass, and higher sensitivity to dynamic loading. Conventional deterministic design methods often overlook the natural variability and short-term amplification effects caused by walking and other intermittent human activities. This study presents a physics-guided, multi-modal data-driven framework for predicting vibration responses and assessing structural reliability under human-induced loading conditions. Monitoring conducted over a 1000-second period shows bounded but dynamically fluctuating acceleration responses ranging between 9.5 m/s² and 10.2 m/s², concentrated around gravitational acceleration. Statistical evaluation indicates an approximately Gaussian distribution with minimal skewness, suggesting stable damping behavior and consistent stiffness properties. Frequency-domain analysis highlights dominant low-frequency components without noticeable resonance amplification within the typical walking frequency range of 1–4 Hz. Rolling mean trends reveal negligible long-term drift, confirming stable structural performance. The framework combines acceleration, strain, and environmental measurements to improve predictive capability and reliability evaluation. Correlation analysis shows that strain measurements strengthen response prediction, while temperature has a comparatively minor effect. Reliability assessment based on a probabilistic threshold model (μ + 2σ) identifies very few exceedance events, indicating a high reliability level and compliance with serviceability requirements. By integrating probabilistic modeling with physics-based structural constraints, the approach enables more accurate forecasting of vibration behavior and real-time reliability assessment. Overall, the results demonstrate improved prediction performance and provide a practical methodology for monitoring and evaluating serviceability in modern lightweight and composite floor systems.},
      keywords = {Human-Induced Vibrations; Lightweight Floor Systems; Composite Structures; Multi-Modal Data Fusion; Stochastic Modeling; Reliability Assessment; Structural Health Monitoring; Serviceability Analysis.},
      month = {September},
      doi = {https://doi.org/10.64388/IREV10I3-1722945}
  }