Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management

Abstract

Making appropriate decisions is indeed a key factor to help companies facing challenges from supply chains nowadays. In this paper, we propose two datadriven approaches that allow making better decisions in supply chain management. In particular, we suggest a Long Short Term Memory (LSTM) networkbased method for forecasting multivariate time series data and an LSTM Autoencoder network-based method combined with a one-class support vector machine algorithm for detecting anomalies in sales. Unlike other approaches, we recommend combining external and internal company data sources for the purpose of enhancing the performance of forecasting algorithms using multivariate LSTM with the optimal hyperparameters. In addition, we also propose a method to optimize hyperparameters for hybrid algorithms for detecting anomalies in time series data. The proposed approaches will be applied to both benchmarking datasets and real data in fashion retail. The obtained results show that the LSTM Autoencoder based method leads to better performance for anomaly detection compared to the LSTM based method suggested in a previous study.The proposed forecasting method for multivariate time series data also performs better some other methods based on a dataset provided by NASA.

Web Traffic Anomaly Detection using Stacked Long Short-Term Memory

“…ℎ 𝑡 = 𝑜 𝑡 tanh(𝑐 𝑡 ), (6) where 𝑊 and 𝑏 are weights and biases, 𝑥 are inputs,𝑓, 𝑖, and 𝑜 are forget gates, input gates, and output gates, 𝑐̃ is a candidate of cell state, 𝑐 is cell state, and ℎ is hidden output [11]. The information carried by the previous cell state 𝑐 𝑡−1 will be decided to be forgotten or passed by multiplying the values of 𝑐 𝑡−1 and the forget gate 𝑓 𝑡 by the inputs 𝑥 𝑡 and ℎ 𝑡−1 .…”

“…To form a forecast model, there is no specific value for the window size [11]. We use several window sizes i.e.…”

“…LSTM is an effective model for studying sequential data with a long-term relationship and represents the relationship between current events and past events [11]. Therefore, in this study, anomaly detection was carried out on web traffic data using a stacked LSTM by developing the method used in Braei and Wagner's research [10].…”

Evaluation of interpretability methods for multivariate time series forecasting

“…On the other hand, the research on interpretability of time series regression models mainly focuses on intrinsic explainability [22], and the absence of proper evaluation metrics for local explanation methods can be deemed as one possible reason for the lack of studies on interpreting time series regression models. Interpreting time series regression models is equally important to those of time series classification, as these are highly relevant in many areas including electricity load [32] and wind speed [31] forecasting, anomaly detection [35], spectrum occupancy prediction [38], sales forecasting [20], and more recently, for COVID-19 spread forecasting [11,18]. Considering the large number of real-world applications, increased interpretability of these models can be highly important for practitioners in many domains.…”

Aircraft Track Anomaly Detection Based on MOD-Bi-LSTM

“…Existing track anomaly detection methods include methods based on the Multidimensional Local Outlier Factor [8], extracting global track features [17], classifiers [18,19], track segment similarity detection [20,21], etc. In recent years, with the advancement of deep learning methods in the big data detection field, anomaly detection methods of deep learning have come to include an anomaly detection algorithm based on multi-layer convolution neural network interactive visualization [22], a Longshort Time Memory Network anomaly detection algorithm based on the encoder-decoder framework [23], a novel intrusion detector based on deep learning hybrid methods [24], forecasting and anomaly detection approaches using LSTM and LSTM autoencoder techniques with the applications in supply chain management [25], a new method for anomaly detection of seismic preprecursor data based on LSTM-RNN [26], a deep learning-based hybrid intelligent intrusion detection system [27], deep learning for anomaly detection [28], a track anomaly detection algorithm based on the Bidirectional Long-short Time Memory Network [9], etc. In addition, Ruff et al published a review on deep and shallow anomaly detection [29].…”