Anomaly Detection in Road-Tanker Fuel Transport: A Deep Learning and Simulation Study
Keywords:
Fuel theft detection, BiLSTM, Attention mechanism, LSTM Autoencoder, Random ForestAbstract
Fuel theft in road-tanker fleets is challenging to detect due to scarce labels, sensor noise, and distribution shifts. We present a reproducible, simulation-driven framework that combines a configurable trip simulator (geofence, speed, tank level, and outflow with realistic noise) with two detection strategies. An unsupervised LSTM Autoencoder (AE) is trained on normal windows to produce anomaly-sensitive auxiliary features. On top of these, we build: (i) an end-to-end BiLSTM+Attention model (Seq-Attn) that fuses temporal context with auxiliary features, and (ii) a hybrid LSTM+RF approach that encodes each window with an LSTM and classifies the embedding via a Random Forest. We evaluate on three scenarios---Baseline, Class Imbalance, and Distribution Shift---using trip-wise splits and theft-centric metrics (Precision, Recall, F1, PR-AUC, ROC-AUC). Results show both models achieve near-perfect detection in the baseline. Under imbalance and shift, Seq-Attn achieves consistently higher recall with almost perfect precision, while LSTM+RF remains competitive but misses slightly more thefts. These findings suggest using Seq-Attn as the primary detector, with LSTM+RF as a supplementary model for interpretability and ensemble robustness. All code and artifacts are released for end-to-end reproducibility and future validation.
Downloads
References
[1] Bishop, C. M., & Nasrabadi, N. M. (2006). Pattern recognition and machine learning (Vol. 4,
No. 4, p. 738). New York: springer.
[2] Hastie, T., Tibshirani, R., Friedman, J., & Franklin, J. (2005). The elements of statistical
learning: data mining, inference and prediction. The Mathematical Intelligencer, 27(2), 83-85.
[3] Sutton, R. S., & Barto, A. G. (1998). Reinforcement learning: An introduction (Vol. 1, No. 1,
pp. 9—11). Cambridge: MIT press.
[4] Mohammadpour, N., Fozi, M., Ebadzadeh, M. M., Azimi, A., & Kamali, A. (2024). Proximal
Policy Optimization with Adaptive Generalized Advantage Estimate. In Proceedings of the First
International Conference on Machine Learning and Knowledge Discovery (MLKD 2024). (pp.
453—458).
[5] Barbado, A., & Corcho, Ó. (2022). Interpretable machine learning models for predicting and
explaining vehicle fuel consumption anomalies. Engineering Applications of Artificial
Intelligence, 115, 105222.
[6] Mulongo, J., Atemkeng, M., Ansah-Narh, T., Rockefeller, R., Nguegnang, G. M., & Garuti, M.
A. (2020). Anomaly detection in power generation plants using machine learning and neural
networks. Applied Artificial Intelligence, 34(1), 64-79.
[7] Atemkeng, M., Osanyindoro, V., Rockefeller, R., Hamlomo, S., Mulongo, J., Ansah-Narh, T., ...
& Fadja, A. N. (2023). Label assisted autoencoder for anomaly detection in power generation
plants. arXiv preprint arXiv:2302.02896.
[8] Shi, J., Gao, Y., Gu, D., Li, Y., & Chen, K. (2023). A novel approach to detect electricity theft
based on conv-attentional Transformer Neural Network. International Journal of Electrical
Power & Energy Systems, 145, 108642.
[9] Lilhore, U. K., Dalal, S., Radulescu, M., & Barbulescu, M. (2025). Smart grid stability
prediction model using two-way attention based hybrid deep learning and MPSO. Energy
Exploration & Exploitation, 43(1), 142-168.
[10] Zhang, X., Shi, J., Yang, M., Huang, X., Usmani, A. S., Chen, G., ... & Li, J. (2023). Real-time
pipeline leak detection and localization using an attention-based LSTM approach. Process
Safety and Environmental Protection, 174, 460-472.
[11] Khattak, A., Bukhsh, R., Aslam, S., Yafoz, A., Alghushairy, O., & Alsini, R. (2022). A hybrid
deep learning-based model for detection of electricity losses using big data in power
systems. Sustainability, 14(20), 13627.
[12] Collier, C., & Guha, K. (2025). Lightweight LSTM Model for Energy Theft Detection via Input
Data Reduction. arXiv preprint arXiv:2507.02872.
[13] Hashmi, A., Barukab, O. M., & Hamza Osman, A. (2024). A hybrid feature weighted attention
based deep learning approach for an intrusion detection system using the random forest
algorithm. Plos one, 19(5), e0302294.
[14] Arapidis, E., Temenos, N., Giagkos, D., Rallis, I., Kalogeras, D., Papadakis, N., ... & C.
Messinis, S. (2024, June). Zeekflow+: A Deep LSTM Autoencoder with Integrated Random
Forest Classifier for Binary and Multi-class Classification in Network Traffic Data.
In Proceedings of the 17th International Conference on PErvasive Technologies Related to
Assistive Environments (pp. 613-618).
[15] Takele, A. K., & Villányi, B. (2022, May). Anomaly detection using hybrid learning for
industrial IoT. In 2022 IEEE 2nd Conference on Information Technology and Data Science
(CITDS) (pp. 262-266).
