Detecting Rough Handling Events in Parcel Transportation using Unsupervised Learning
Article Sidebar
Main Article Content
Abstract
The growth of e-commerce has intensified the impact of parcel damage caused by mishandling during transit. The study proposes an unsupervised anomaly detection to identify real time abnormal events during parcel transit using low cost inertial sensor data from a consumer grade smartphone embedded within a parcel. Data was collected for manual parcel handling, motorcycle, and vehicular transport under ‘normal’ and ‘rough’ handling conditions. Data collected in ‘normal’ conditions were used to train IForest and LSTM Autoencoder unsupervised models, and they were evaluated on parcel handling data collected on ‘rough’ conditions. Precision, recall, F1-Score, and False Positives Rate (FPR) metrics were used to assess model performance, and results indicated that both models effectively detect anomalies for manual and motorcycle handling, but report lower performance in vehicular transportation. The study contributes to the body of knowledge by demonstrating the feasibility of using smartphone-based inertial sensing for parcel monitoring and by providing a comparative evaluation of classical (IForest) and deep learning (LSTM Autoencoder) anomaly detection approaches across multiple transport modes. Overall, IForest reported more robust and consistent performance across heterogeneous transport modes, while LSTM Autoencoder achieved a lower FPR in contexts with sustained temporal irregularities. Both highlight the feasibility of utilizing unsupervised anomaly detection for parcel monitoring in complex freight transportation environments.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
[1] S. Jakovlev et al., “Detection of Physical Impacts of Shipping Containers during Handling Operations using the Impact Detection Methodology,” J. Mar. Sci. Eng., vol. 10, no. 9, pp. 1250, 2022.
[2] Z. Kavaliauskas, I. Sajev, G. Blaziunas, G. Gecevicius and S. Kazlauskas, “Development and Investigation of A Smart Impact Detector for Monitoring the Shipment Transport Process,” Appl. Sci., vol. 14, no. 16, pp. 7102, Aug. 2024.
[3] T. Matic, J. Zidar, I. Aleksi and D. Zagar, “Smart Sticker Ultra-Low-Power Shock Detection in the Supply Chain,” Sensors, vol. 22, no. 11, pp. 4003, 2022.
[4] A. T. Almaktoom, D. Gupta, S. A. Alsobhi and K. K. Krishnan, “Analysis of Damage Costs in Supply Chain Systems,” Int. J. Industr. and Syst. Eng., vol. 28, no. 1, pp. 70-98, 2018.
[5] X. Yang, X. Dai and Z. Liu, “The Impact of Return Shipping Insurance on A Retailer Based on Restricting Rights,” J. Theor. and Appl. Electron. Commer. Res., vol. 18, no. 1, pp. 706–724, 2023.
[6] S. Zhao, “Managing Service-Level Returns in E-Commerce: Joint Pricing, Delivery Time, and Handling Strategy Decisions,” J. Theor. and Appl. Electron. Commer. Res., vol. 20, no. 4, pp. 282, 2025.
[7] “Total Logistics Report 2023,” Transp. Intell. Res., 2023.
[8] P. Panjee, V. Kaewchueaknang and S. Amornsawadwatana, “A Systematic Literature Review of Cargo Loss Risks in Road Transportation: Impacts and Future Directions,” Safety, vol. 11, no. 1, pp. 20, 2025.
[9] X. Yu et al., “Applications, Challenges, and Prospects of AI-Driven Intelligent Packaging Technologies in The Food Supply Chain under Industry 4.0,” Trends Food Sci. Technol., vol. 165, pp. 105367, 2025.
[10] Y. Zhang, W. Yang, H. Huang, C. Duan and L. Zhu, “A Lightweight Framework for Recognizing Rough Handling Behaviours in Parcel Logistics,” Measurement, vol. 259, pp. 119693, 2026.
[11] T. P. da Costa et al., “A Systematic Review of Real-Time Monitoring Technologies and Its Potential Application to Reduce Food Loss and Waste: Key Elements of Food Supply Chains and IoT Technologies,” Sustainability, vol. 15, no. 1, pp. 614, 2022.
[12] M. Feng, Y. Fukuda, M. Mizuta and E. Ozer, “Citizen Sensors for SHM: Use of Accelerometer Data from Smartphones,” Sensors, vol. 15, no. 2, pp. 2980–2998, 2015.
[13] H. Sarmadi, A. Entezami, K. V. Yuen and B. Behkamal, “Review on Smartphone Sensing Technology for Structural Health Monitoring,” Measurement, vol. 223, pp 113716, 2023.
[14] E. A. Martinez-Ríos, M. R. Bustamante-Bello and L. A. Arce-Sáenz, “A Review of Road Surface Anomaly Detection and Classification Systems Based on Vibration-Based Techniques,” Appl. Sci., vol. 12, no. 19, pp. 9413, 2022.
[15] A. Khandakar et al., “Harnessing Smartphone Sensors for Enhanced Road Safety: A Comprehensive Dataset and Review,” Sci. Data, vol. 12, no. 1, pp. 418, 2025.
[16] F. T. Liu, K. M. Ting and Z. H. Zhou, “Isolation-Based Anomaly Detection,” ACM Trans. Knowl. Discov. Data, vol. 6, no. 1, pp. 1–39, 2012.
[17] W. S. Al Farizi, I. Hidayah and M. N. Rizal, “IForest Based Anomaly Detection: A Systematic Literature Review,” in 2021 8th Int. Conf. Informat. Technol., Comput. and Electr. Eng., pp. 118–122, 2021.
[18] M. Wang et al., “Anomaly Detection in Multidimensional Time Series for Water Injection Pump Operations Based on LSTMA-AE and Mechanism Constraints,” Sci. Rep., vol. 15, no. 1, pp. 2020, 2025.
[19] Y. Lee, C. Park, N. Kim, J. Ahn and J. Jeong, “LSTM-Autoencoder Based Anomaly Detection Using Vibration Data of Wind Turbines,” Sensors, vol. 24, no. 9, pp. 2833, 2024.
[20] F. Rewicki, J. Denzler and J. Niebling, “Is It Worth It? Comparing Six Deep and Classical Methods for Unsupervised Anomaly Detection in Time Series,” Appl. Sci., vol. 13, no. 3, p. 1778, 2023.
[21] F. Wang et al., “A Survey of Deep Anomaly Detection in Multivariate Time Series: Taxonomy, Applications, and Directions,” Sensors, vol. 25, no. 1, pp. 190, 2025.
[22] “phyphox – Physical Phone Experiments,” RWTH Aachen University. [Available Online on 28 November 2025] https://phyphox.org.
[23] J. Zipfel et al., “Anomaly Detection for Industrial Quality Assurance: A Comparative Evaluation of Unsupervised Deep Learning Models,” Comput. Ind. Eng., vol. 177, pp. 109045, 2023.