DEEP LEARNING BASED DETECTION OF ANOMALOUS INVENTORY FLUCTUATIONS IN INDUSTRIAL SUPPLY CHAINS
Keywords:
Industrial supply chain, Inventory anomaly detection, Deep learning, Time seriesAbstract
Fluctuations in inventories across industrial supply chains are caused by interplay between demand, ordering, replenishment, and production which makes it challenging to detect anomalies with just one threshold or simple distributional rules. In this paper we introduce an anomaly detection approach based on four synchronized time series (inventory, demand, ordering, and replenishment). Our model captures normal operating patterns through sliding windows and computes the difference between actual and estimated inventory level as an anomaly score. We tested our model over a 240-step simulated sequence including sudden inventory drops and persistent inventory surpluses; for this sequence the multivariate model obtained an F1 score of 0.897 and an AUC of 0.951, surpassing Isolation Forest, LSTM, GRU, and baseline Transformer models. Detected anomalies matched the injected anomaly events. Results indicate potential use of multivariate time-series modeling for inventory anomaly screening, however, further validation is needed on real-world enterprise data.References
[1] Cui L, Chen Y, Deng J, et al. A novel attLSTM framework combining the attention mechanism and bidirectional LSTM for demand forecasting. Expert Systems with Applications, 2024, 254: 124409.
[2] Douaioui K, Oucheikh R, Benmoussa O, et al. Machine learning and deep learning models for demand forecasting in supply chain management: a critical review. Applied System Innovation, 2024, 7(5): 93.
[3] Guo W, Ren Z, Du W, et al. TFT-MPIR: an end-to-end multi-period inventory replenishment strategy based on temporal fusion transformer. Expert Systems with Applications, 2025, 261: 125464.
[4] Kang H, Kang P. Transformer-based multivariate time series anomaly detection using inter-variable attention mechanism. Knowledge-Based Systems, 2024, 290: 111507.
[5] Lim B, Arik S O, Loeff N, et al. Temporal Fusion Transformers for interpretable multi-horizon time series forecasting. International Journal of Forecasting, 2021, 37(4): 1748-1764.
[6] Seyedan M, Mafakheri F, Wang C. Order-up-to-level inventory optimization model using time-series demand forecasting with ensemble deep learning. Supply Chain Analytics, 2023, 3: 100024.
[7] Tuli S, Casale G, Jennings N R. TranAD: deep Transformer networks for anomaly detection in multivariate time series data. Proceedings of the VLDB Endowment, 2022, 15(6): 1201-1214.
[8] Xu J, Wu H, Wang J, et al. Anomaly Transformer: time series anomaly detection with association discrepancy. International Conference on Learning Representations, 2022.
[9] Yu C, She C, Li J, et al. Research on measuring and predicting the carbon footprint of supply chain throughout its life cycle based on LSTM. Industrial Engineering Journal, 2024, 27(5): 161-171.
[10] Zhou Z, Jiang J. Research on demand forecasting of rarely-used spare parts based on deep artificial neural network. Journal of Hebei University of Water Resources and Electric Engineering, 2021, 31(4): 60-65.