Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy

ARĞUN, Halil; ALPTEKİN, S. Emre

doi:10.55730/1300-0632.3977

Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy

Halil ARĞUN, (Galatasaray Üniversitesi, Mühendislik ve Teknoloji Fakültesi, Endüstri Mühendisliği Bölümü, İstanbul, Türkiye)

S. Emre ALPTEKİN (Galatasaray Üniversitesi, Mühendislik ve Teknoloji Fakültesi, Endüstri Mühendisliği Bölümü, İstanbul, Türkiye)

Turkish Journal of Electrical Engineering and Computer Sciences

1 0

Yıl: 2023 Cilt: 31 Sayı: 1 Sayfa Aralığı: 163 - 179 Metin Dili: İngilizce DOI: 10.55730/1300-0632.3977 İndeks Tarihi: 16-05-2023

Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy

Öz:

Retail companies monitor inventory stock levels regularly and manage them based on forecasted sales to sustain their market position. Inventory accuracy, defined as the difference between the warehouse stock records and the actual inventory, is critical for preventing stockouts and shortages. The root causes of inventory inaccuracy are the employee or customer theft, product damage or spoilage, and wrong shipments. In this paper, we aim at detecting inaccurate stocks of one of Turkey’s largest supermarket chain using the variational autoencoder (VAE), which is an unsupervised learning method. Based on the findings, we showed that VAE is able to model the underlying probability distribution of data, regenerate the pattern from time series data, and detect anomalies. Hence, it reduces time and effort to manually label the inaccuracy in data. Since the distribution of inventory data depends on selected product/product categories, we had to use a parametric approach to handle potential differences. For individual products, we built univariate time series, whereas for product categories we built multivariate time series. The experimental results show that the proposed approaches can detect anomalies both in the low and high inventory quantities.

Anahtar Kelime: Inventory record inaccuracy variational autoencoder anomaly detection time series data

Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık

[1] Stevenson WJ. Operations Management 14th Edition. McGraw-Hill, 2021.
[2] Chuang HHC, Oliva R, Liu S. On-shelf availability, retail performance, and external audits: a field experiment. Production and Operations Management 2016; 25: 935–951. https://doi.org/10.1111/poms.12519
[3] Natsvlishvili E, Lomtadze E, Khatiashvili N. ERP system implementation challenges in Georgian Medium-sized enterprises, retail sector. PhD, Ilia State University, USA, 2020.
[4] Shabani A, Maroti G, de Leeuw S, Dullaert W. Inventory record inaccuracy and store-level performance. Interna- tional Journal of Production Economics 2021; 235: 1-16. https://doi.org/10.1016/j.ijpe.2021.108111
[5] Sarker IH. Machine learning: Algorithms, Real-World Applications and Research Directions. SN Computer Science 2021; 2 (160): 1–21. https://doi.org/10.1007/s42979-021-00592-x
[6] Aggarwal CC. An Introduction to Outlier Analysis. In: Outlier Analysis. New, York, NY: Springer New York, pp. 1-40, 2013. doi:10.1007/978-1-4614-6396-2
[7] Pang G, Shen C, Cao L, Hengel AVD. 2021. Deep Learning for Anomaly Detection: A Review. ACM Computing Surveys 2022; 54 (2): 1-38. https://doi.org/10.1145/3439950
[8] Tayeh T, Aburakhia S, Myers R, Shami A. Distance-Based Anomaly Detection for Industrial Surfaces Using Triplet Networks. In: 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), Vancouver, VC, Canada; 2020, pp. 0372-0377. doi: 10.1109/IEMCON51383.2020.9284921
[9] Lou C, Zhao H. Local Density-Based Anomaly Detection in Hyperspectral Image. Journal of Applied Remote Sensing. 2015; 9 (1): 095070. https://doi.org/10.1117/1.JRS.9.095070
[10] Kriegel HP, Schubert M, Zimek A. Angle-based Outlier Detection in High-Dimensional Data. In: Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD’08); New York, NY, USA; 2008, pp. 444–452. https://doi.org/10.1145/1401890.1401946
[11] Zhang CK, Li SZ, Zhang H, Chen Y. VELC: A New Variational AutoEncoder Based Model for Time Series Anomaly Detection. arXiv:1907.01702; 2020. https://doi.org/10.48550/arXiv.1907.01702
[12] Hinton, GE, Salakhutdinov RR. Reducing the dimensionality of data with neural networks. Science 2006; 313 (5786): 504-507. https://doi.org/10.1126/science.1127647
[13] Kingma DP, Welling M. Auto-Encoding Variational Bayes. 2013. https://doi.org/10.48550/arXiv.1312.6114
[14] An J, Cho S. Variational Autoencoder Based Anomaly Detection Using Reconstruction Probability. Special Lecture on IE 2015; 2 (1): 1–18.
[15] Chehrazi N. Impacts of Inventory Record Inaccuracy on Retailers’ Internal Operations. Available at SSRN; 2020. https://dx.doi.org/10.2139/ssrn.3637031
[16] Zadeh AH, Sharda R, Kasiri N. Inventory record inaccuracy due to theft in production-inventory systems. The International Journal of Advanced Manufacturing 2016; 83: 623–631. https://doi.org/10.1007/s00170-015-7433-3
[17] Kok AG, Shang KH. Evaluation of cycle-count policies for supply chains with inventory inaccuracy and implications on RFID investments. European Journal of Operational Research 2014; 237(1): 91–105. https://doi.org/10.1016/j.ejor.2014.01.052
[18] Zhang Y, Chen Y, Wang J, Pan Z. Unsupervised deep anomaly detection for multi-sensor time-series signals. IEEE Transactions on Knowledge and Data Engineering, 2021; 1-14. https://doi.org/10.1109/TKDE.2021.3102110
[19] Tokovarov M, Karczmarek P. A probabilistic generalization of isolation forest. Information Sciences 2022; 584: 433-449. https://doi.org/10.1016/j.ins.2021.10.075
[20] Zhang R, Zhang S, Muthuraman S, Jiang J. One class support vector machine for anomaly detection in the communication network performance data. In: Proceedings of the 5th Conference on Applied Elec- tromagnetics, Wireless and Optical Communications; Tenerife Canary Islands, Spain; 2007. pp. 31–37. https://doi.org/10.5555/1503549.1503556
[21] Filonov P, Kitashov F, Lavrentyev A. RNN-based Early Cyber-Attack Detection for the Tennessee Eastman Process. arXiv preprint arXiv:1709.02232; 2017. https://doi.org/10.48550/arXiv.1709.02232
[22] Lindemann B, Jazdi N, Weyrich M. Anomaly Detection and Prediction in Discrete Manufacturing Based on Cooperative LSTM Networks. In: IEEE 16th International Conference on Automation Science and Engineering (CASE); Online Zoom Meeting; 2020. pp. 1003–1010. https://doi.org/10.1109/CASE48305.2020.9216855
[23] Malhotra P, Ramakrishnan A, Anand G, Vig L, Agarwal P et al. LSTM-based Encoder-Decoder for Multi-Sensor Anomaly Detection. arXiv preprint arXiv:1607.00148; 2016. https://doi.org/10.48550/arXiv.1607.00148
[24] Chakraborty D, Elzarka H. Advanced Machine Learning Techniques for Building Performance Simula- tion: A Comparative Analysis. Journal of Building Performance Simulation 2018; 12 (2): 193-207. https://doi.org/10.1080/19401493.2018.1498538
[25] Garcia S, Luengo J, Herrera F. Data Preparation Basic Models. In: Data Preprocessing in Data Mining. Switzerland: Springer Cham, 2015. pp. 39-57. https://doi.org/10.1007/978-3-319-10247-4
[26] Nguyen H, Tran KP, Thomassey S, Hamad M. Forecasting and Anomaly Detection Approaches using LSTM and LSTM Autoencoder Techniques with the Applications in Supply Chain Management. International Journal of Information Management 2021; 51: 1-13. https://doi.org/10.1016/j.ijinfomgt.2020.102282
[27] Metlapalli AC, Muthusamy T, Battula BP. Classification of Social Media Text Spam Using VAE-CNN and LSTM Model. Ingénierie des Systèmes d’Information 2020; 25 (6): 747-753. https://doi.org/10.18280/isi.250605
[28] Kokoska S, Zwillinger D. CRC Standard Probability and Statistics Tables and Formulae: CRC Press, 2000.
[29] Efron B. Missing Data, Imputation, and the Bootstrap. Journal of the American Statistical Association 1994; 89 (426): 463-475. https://doi.org/10.2307/2290846
[30] Chen, Tingting and Liu, Xueping and Xia, Bizhong and Wang, Wei and Lai, Yongzhi IEEE Access 2020; 47072– 47081
[31] Akash D, Kanishk G, Deepak Kumar S. Chapter 1 - An introduction to deep learning applications in biometric recognition. In: Hybrid Computational Intelligence for Pattern Analysis, Trends in Deep Learning Methodologies. Academic Press, 2021. pp. 1-36. https://doi.org/10.1016/B978-0-12-822226-3.00001-5

APA	ARĞUN H, ALPTEKİN S (2023). Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. , 163 - 179. 10.55730/1300-0632.3977
Chicago	ARĞUN Halil,ALPTEKİN S. Emre Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. (2023): 163 - 179. 10.55730/1300-0632.3977
MLA	ARĞUN Halil,ALPTEKİN S. Emre Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. , 2023, ss.163 - 179. 10.55730/1300-0632.3977
AMA	ARĞUN H,ALPTEKİN S Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. . 2023; 163 - 179. 10.55730/1300-0632.3977
Vancouver	ARĞUN H,ALPTEKİN S Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. . 2023; 163 - 179. 10.55730/1300-0632.3977
IEEE	ARĞUN H,ALPTEKİN S "Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy." , ss.163 - 179, 2023. 10.55730/1300-0632.3977
ISNAD	ARĞUN, Halil - ALPTEKİN, S. Emre. "Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy". (2023), 163-179. https://doi.org/10.55730/1300-0632.3977

APA	ARĞUN H, ALPTEKİN S (2023). Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. Turkish Journal of Electrical Engineering and Computer Sciences, 31(1), 163 - 179. 10.55730/1300-0632.3977
Chicago	ARĞUN Halil,ALPTEKİN S. Emre Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. Turkish Journal of Electrical Engineering and Computer Sciences 31, no.1 (2023): 163 - 179. 10.55730/1300-0632.3977
MLA	ARĞUN Halil,ALPTEKİN S. Emre Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. Turkish Journal of Electrical Engineering and Computer Sciences, vol.31, no.1, 2023, ss.163 - 179. 10.55730/1300-0632.3977
AMA	ARĞUN H,ALPTEKİN S Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. Turkish Journal of Electrical Engineering and Computer Sciences. 2023; 31(1): 163 - 179. 10.55730/1300-0632.3977
Vancouver	ARĞUN H,ALPTEKİN S Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy. Turkish Journal of Electrical Engineering and Computer Sciences. 2023; 31(1): 163 - 179. 10.55730/1300-0632.3977
IEEE	ARĞUN H,ALPTEKİN S "Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy." Turkish Journal of Electrical Engineering and Computer Sciences, 31, ss.163 - 179, 2023. 10.55730/1300-0632.3977
ISNAD	ARĞUN, Halil - ALPTEKİN, S. Emre. "Variational autoencoder-based anomaly detection in time series data for inventory record inaccuracy". Turkish Journal of Electrical Engineering and Computer Sciences 31/1 (2023), 163-179. https://doi.org/10.55730/1300-0632.3977