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

In recent years, the rapid development and wide application of cutting-edge technologies have profoundly impacted the industrial production area, leading to the idea of ​​Smart Manufacturing.

Smart manufacturing promotes the automation process with less human intervention as well as the flexibility that enables it to detect system failure early and automate the system. The supply chain is a chain or a process that starts from raw materials until the completed product or service reaches the final consumer.
 In Smart Manufacturing, digital supply chains with technologies in terms of artificial intelligence (AI), big data analytics, and sensors in the Internet of Things (IoT) are reshaping supply chains and generating novel opportunities to boost operational efficiency and lower costs.
Understanding the role and essence of supply chain management in smart manufacturing, Dr. Nguyen Huu Du (Director of IAD) and colleagues in France researched and published a scientific article: Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management.

This study was conducted in 2019 when Dr. Nguyen Huu Du was doing post-doctoral research under the guidance of Assoc. Dr. Tran Kim Phuc at ENSAIT, France. The research was carried out in collaboration with a large French company in the fashion industry, apart from academic contributions, the study also proved the application of artificial intelligence in optimizing the digital technical supply chain at the company.
The article was published in the International Journal of Information Management on December 16th, 2020. Currently, this article is cited 31 times from other research papers. In 2021, this journal has IF > 14 (Scopus) and H index 114, is Q1 with ranking 7/202 (top 4%)  of SCIE journals in the AI field. (Read more information about International Journal Of Information Management). Based on these impressive achievements, the article is being nominated for the IAD Best Article Award 2020.

  • Two data-driven approaches are proposed to enhance decision-making better in the supply chain.
  • A multivariate time series forecasting is performed with a Long Short Term Memory (LSTM) network-based method.
  • LSTM Autoencoder network-based method combined with a one-class support vector machine.
  • The proposed approach is implemented to both benchmarking and real datasets.

This article may help us broaden professional knowledge in implementing AI to supply chain, so that let’s read and explore it here