International Chair in Data Science and Explainable Artificial Intelligence


About International Chair in DS & XAI

Towards the goal of developing further studies about Data Science and Explainable Artificial Intelligence, the International Chair in Data Science and Explainable Artificial Intelligence was founded in September 2018. We have aimed to develop a roadmap that starts from Industry 4.0 to reach Industry 5.0 and beyond which drives sustainability since 2018. Studies about Industry 5.0 will focus on combining human creativity and craftsmanship with the speed, productivity, and consistency of AI systems.


The Chair aims to develop new theoretically and computationally solid techniques of Data Science and Explainable Artificial Intelligence with a wide class of applications in Manufacturing, Healthcare, Business, and Finance.


International Chair in Data Science and Explainable Artificial Intelligence will establish a Joint Research Insitute of Vietnam & France in the next 10 years. Particularly, the main organizations are the University of Lille, France, and Dong A University (UDA), Vietnam, in which IAD is the key organization and a representative of UDA. 

Research topics

  • Embedded Federated Learning
  • Explainability of AI-based prediction
  • Trustworthy AI systems 
  • Explainable Anomaly Detection
  • Reinforcement Learning
  • Cybersecurity 
  • Decision Support Systems
  • Fairness of AI algorithms
  • Privacy-preserving data analytics
  • Auditing AI systems
  • Distributed/online optimization algorithms
  • Coupling optimization/simulation with AI
  • Data-driven supply chain network design
  • Wearable technologies
  • Smart Healthcare
  • Smart Manufacturing
  • Data Science
  • Inverse problems in medical imaging

Head of the International Chair

 Senior Assoc. Prof. Kim Phuc Tran

  University of Lille, ENSAIT, GEMTEX, France

-  Google Scholar:
-  Personal webpage:


Kim Phuc Tran is currently a Senior Associate Professor (Maître de Conférences HDR, equivalent to a UK Reader) of Artificial Intelligence and Data Science at the ENSAIT and the GEMTEX laboratory, University of Lille, France. He received an Engineer's degree and a Master of Engineering degree in Automated Manufacturing. He obtained a Ph.D. in Automation and Applied Informatics at the University of Nantes, and an HDR (Doctor of Science or Dr. habil.) in Computer Science and Automation at the University of Lille, France. His research deals with Real-time Anomaly Detection with Machine Learning with applications, Decision Support Systems with Artificial Intelligence, and Enabling Smart Manufacturing with IIoT, Federated learning, and Edge computing. He has published more than 64 papers in peer-reviewed international journals and proceedings of international conferences. He edited 3 books with Springer Nature and Taylor & Francis. He is the Associate Editor, Editorial Board Member, and Guest Editor for several international journals such as IEEE Transactions on Intelligent Transportation Systems and Engineering Applications of Artificial Intelligence. He has supervised 9 Ph.D. students and 3 Postdocs. In addition, as the project coordinator (PI), he is conducting 1 regional research project about Healthcare Systems with Federated Learning. He has been or is involved (co-PI or member) in 8 national and European projects. He is an expert and evaluator for the Public Service of Wallonia (SPW-EER), Belgium and the Natural Sciences and Engineering Research Council of Canada. He received the Award for Scientific Excellence (Prime d’Encadrement Doctoral et de Recherche) given by the Ministry of Higher Education, Research and Innovation, France for 4 years from 2021 to 2025 in recognition of his outstanding scientific achievements.

From 2017 until now, he has been the Senior Scientific Advisor at Dong A University and the International Research Institute for Artificial Intelligence and Data Science (IAD), Danang, Vietnam where he has held the International Chair in Data Science and Explainable Artificial Intelligence.

Local Chair

MBA. Kim Duc Tran

Senior members 

Prof. Cédric Heuchenne (HEC, University of Liège, Belgium)
Assist. Prof. Athanasios ‘Sakis’ Rakitzis (University of Piraeus, Greece, University of the Aegean)


Dr. Huu Du Nguyen (Hanoi University of Science and Technology, Vietnam)
Dr. Quoc Thong Nguyen
Assist. Prof. Hai Canh Vu (The University of Technology of Compiègne, Compiègne, France)
Assist. Prof. Thi Hien Nguyen (Laboratoire AGM, UMR CNRS 8088, CY Cergy Paris Université, 95000 Cergy, France)
Dr. Duc Phong Nguyen
PhD. Student Phuong Hanh Tran (HEC, University of Liège, Belgium)
PhD. Student Thi Thuy Van Nguyen  (HEC, University of Liège, Belgium, ENSAIT, GEMTEX, the University of Lille France, IAD, Dong A University, Vietnam)
Engineer Do Thu Ha (University of Lille, ENSAIT, GEMTEX, France, IAD, Dong A University)
- PhD. Student Phuong Bac Ta (Soongsil University, Korea, IAD, Dong A University, Vietnam)
MD. Tho Nguyen
PhD. Student Ali Raza (University of Lille, ENSAIT, GEMTEX, France, School of Computing & Institute of Cyber Security for Society (iCSS), University of Kent, UK)
Dr. Adel Ahmadi Nadi (University of Waterloo, Canada, University of Lille, ENSAIT, GEMTEX, France)
Assoc. Prof. Aamir Saghir (Mirpur University of Science and Technology (MUST), Mirpur-10250(AJK), Pakistan)
- Assist. Prof. Robab Afshari (University of Zanjan, Iran)
- Dr. Muhammad Imran (Guangzhou University, China)
- Dr. Fatima Sehar Zaidi (Guangzhou University, China)


1. Journals

 A. In 2023

  1. H. D. Nguyen, H. L. Nguyen,  N. H. Kieu, V. H. Nguyen,  T. H. Truong, & K. P. Tran (2023). Trans-Lighter: A light-weight federated learning-based architecture for Remaining Useful Lifetime predictionComputers in Industry148, 103888.
  2. A. Raza, K. P. Tran, L. Koehl, & S. Li (2023). AnoFed: Adaptive anomaly detection for digital health using transformer-based federated learning and support vector data descriptionEngineering Applications of Artificial Intelligence121, 106051.
  3. A. A. Nadi, B. S. Gildeh, J. Kazempoor, K. D. Tran, & K. P. Tran (2023). Cost-effective optimization strategies and sampling plan for Weibull quantiles under type-II censoringApplied Mathematical Modelling116, 16-31.
  4. F.S. Zaidi, H.L Dai, M.  Imran, K.P. Tran (2023). Analyzing Abnormal Pattern of Hotelling T2 Control Chart for Compositional Data using Artificial Neural Networks, Computers & Industrial Engineering, 109254.
  5. F.S. Zaidi, H.L Dai, M.  Imran, K.P. Tran,(2023). Monitoring Autocorrelated Compositional Data Vectors using an Enhanced Residuals Hotelling T2 Control Chart, Computers & Industrial Engineering.
  6.  A. Saghir, X. Hu, K.P. Tran, Z. Song,(2023). Optimal design and evaluation of adaptive EWMA monitoring schemes for Inverse Maxwell distribution, Computers & Industrial Engineering.

 B. In 2022

  1. H. D. Nguyen, K. P. Tran, S.Thomassey & M. Hamad (2021). Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain managementInternational Journal of Information Management57, 102282.
  2. H. D. Nguyen, A. A. Nadi, K. D. Tran, P. Castagliola, G. Celano, & K. P. Tran (2022). The Shewhart-type RZ control chart for monitoring the ratio of autocorrelated variablesInternational Journal of Production Research, 1-26.
  3. H. T. Truong, B. P. Ta, Q. A. Le, D. M. Nguyen, C. T. Le, H. X. Nguyen, H.T. Do, H.T. Nguyen & K. P. Tran (2022). Light-weight federated learning-based anomaly detection for time-series data in industrial control systemsComputers in Industry140, 103692.
  4. A. Raza, K. P. Tran, L. Koehl & S. Li (2022). Designing ECG monitoring healthcare system with federated transfer learning and explainable AIKnowledge-Based Systems236, 107763.
  5. Z. He, K. P. Tran, S. Thomassey, X. Zeng, J. Xu & C. Yi (2022). Multi-objective optimization of the textile manufacturing process using deep-Q-network based multi-agent reinforcement learningJournal of Manufacturing Systems62, 939-949.
  6.  A. Raza, S. Li , K. P. Tran, L. Koehl, S. Li & K. Ludovic ​(2022). Detection of Poisoning Attacks with Anomaly Detection in Federated Learning for Healthcare Applications: A Machine Learning ApproacharXiv preprint arXiv:2207.08486
  7.   R. Afshari, A. A. Nadi, A. Johannssen, N. Chukhrova & K. P. Tran (2022). The effects of measurement errors on estimating and assessing the multivariate process capability with imprecise characteristicComputers & Industrial Engineering, 108563.

 C. In 2021

  1. T. H. Truong, P. B. Ta, M. L. Dao, D. L. Tran, M. D. Nguyen, A.Q. Le, T. C. Le, D. T. Bui &  K. P. Tran (2021). Detecting cyberattacks using anomaly detection in industrial control systems: A Federated Learning approachComputers in Industry132, 103509.
  2. K. D. Tran, A. A. Nadi, T. H. Nguyen & K. P. Tran (2021). One-sided Shewhart control charts for monitoring the ratio of two normal variables in short production runsJournal of Manufacturing Processes69, 273-289.
  3. H. D. Nguyen, K. P. Tran & K. D. Tran (2021). The effect of measurement errors on the performance of the Exponentially Weighted Moving Average control charts for the Ratio of Two Normally Distributed VariablesEuropean Journal of Operational Research293(1), 203-218.
  4. Z. He, K. P. Tran, S. Thomassey, X.Zeng, J.Xu & C. Yi (2021). A deep reinforcement learning based multi-criteria decision support system for optimizing textile chemical processComputers in Industry125, 103373.
  5. T. H. Truong, P. B. Ta, M. L. Dao, T. D. Bui,  D. L. Tran &  K. P. Tran (2021). Lockedge: Low-complexity cyberattack detection in iot edge computingIEEE Access9, 29696-29710.
  6. Q. T. Nguyen, V. Giner-Bosch, K. D. Tran, C. Heuchenne & K. P. Tran (2021). One-sided variable sampling interval EWMA control charts for monitoring the multivariate coefficient of variation in the presence of measurement errorsThe International Journal of Advanced Manufacturing Technology115(5), 1821-1851.
  7. A. Raza, K. P. Tran, L. Koehl, S. Li, X. Zeng, & K. Benzaidi (2021). Lightweight Transformer in Federated Setting for Human Activity RecognitionarXiv preprint arXiv:2110.00244.

2. Chapters 

  1. P. H. Tran, A. Ahmadi Nadi, T. H. Nguyen, K. D. Tran & K. P. Tran (2022). Application of Machine Learning in Statistical Process Control Charts: A Survey and Perspective. In Control Charts and Machine Learning for Anomaly Detection in Manufacturing (pp. 7-42). Springer, Cham.
  2. H. D. Nguyen, K. P. Tran, P. Castagliola & F. M. Megahed (2022). Enabling Smart Manufacturing with Artificial Intelligence and Big Data: A Survey and Perspective. In Advanced Manufacturing Methods (pp. 1-26). CRC Press.

3. Conferences 

  1. T.H. Do, X. H. Nguyen, V. H. Nguyen, H. D. Nguyen, T. H. Truong & K. P. Tran (2022, June). Explainable Anomaly Detection for Industrial Control System Cybersecurity. In Proceedings of The IFAC 10th conference on MANUFACTURING MODELING, MANAGEMENT AND CONTROL,  June 22-24, 2022 , Nantes, France.
  2. F. Ouedraogo, C. Heuchenne, Q. T. Nguyen, and H. Tran, "Data-Driven Approach for Credit Card Fraud Detection with Autoencoder and One-Class Classification Techniques", In Proceedings of the Advances in Production Management Systems. Artificial Intelligence for Sustainable and Resilient Production Systems (APMS 2021), Nantes, France, 5–9 Sept, 2021, Proceedings, Part I,
  3. K.P. Tran H. D. Nguyen, Q. T. Nguyen, and W. Chattinnawat (2018) "One-Sided synthetic control charts for monitoring the Coefficient of Variation with Measurement Errors", In processdings of The IEEE International Conference on Industrial Engineering and Engineering Management , 16-19 December, 2018, Bangkok, Thailand.
  4. K.P. Tran, P. Castagliola, T.H. Nguyen and A. Cuzol (2018), "The Efficiency of the VSI Exponentially Weighted Moving Average Median Control Chart", In 24nd ISSAT International Conference on Reliability and Quality in Design, August 2-4, 2018, Toronto, Ontario, Canda.

4. Books


K.P. Tran (Ed.). (2023). Artificial Intelligence for Smart Manufacturing: Methods, Applications, and Challenges (1st ed). Springer. K.P. Tran (Ed.). (2022). Machine Learning and Probabilistic Graphical Models for Decision Support Systems (1st ed.). CRC Press. K.P. Tran (Ed.). (2021). Control Charts and Machine Learning for Anomaly Detection in Manufacturing (1st ed.). Springer.



Embedded Artificial Intelligence Platforms

   Federated learning-based anomaly detection testbed

  Light-weight federated learning-based anomaly detection for time-series data in industrial control systems

Federated learning-based anomaly detection is developed and studied by the collaboration of Future Internet Laboratory – HUSTIAD – UDA, and University of Lille – France.

Main goal is to design an AD system (FATRAF) that has a fast training time and is light weight to accommodate frequent learning update, while still either retaining the same or improving the detection performance in comparison with some existing AD solutions for ICSs in the literature.

Federated learning (FL)-based anomaly detection architecture was proposed a new hybrid solution of Autoencoder, Transformer and Fourier transform operating in a chain to enhance the detection performance, whilst reducing the training time and being more lightweight for more feasible deployment over edge devices of an IIoT-based industrial control system.


Figure 1. Federated learning (FL) – based anomaly detection architecture (FATRAF)

FATRAF comprises two main components:

  • Edge sites: In an IIoT-based smart factory monitored by industrial control systems, there are various manufacturing zones, in which sensor systems are installed to gather readings that signify operating states over time. Subsequently, the time-series data, as the local data, is transmitted wirelessly to edge devices in the vicinity of the corresponding manufacturing zones. Designated to monitor anomalies, these edge devices employ the local data as inputs for training its own local anomaly detection model - ATRAF (AE-Transformer-Fourier learning model). This deployment allows detecting anomalies timely right at the edge sites, making use of the computing capacity of edge devices, and distributing heavy computation tasks that could overload the cloud server.

Figure 2. ATRAF learning model operation during the testing process

  • Cloud Server: The cloud server undertakes two primary functions: system initialization and aggregation of local models sent from different edges. The whole process of model aggregation and global model updating down to all local models are called Federated Learning.

The procedure of FATRAF can be summarized into the following key stages:

  1. System Initialization: At the beginning, the cloud server establishes a global model with specific learning parameters and sending it to each edge device of corresponding edge site.
  2. Local Training: After receiving the initial configuration, by utilizing the on-site data collected from sensors, edge devices conduct a local training process for their Autoencoder blocks. Subsequently, at each edge site, the local training data is fed into the encoder part of the trained Autoencoder model so as to obtain corresponding embeddings or code sequences. These code sequences are then applied to train the Transformer-Fourier block.
  3. Local Model Update: After the local training, each edge device sends the learnable parameters of the Transformer-Fourier block wtr+1 to the cloud server for aggregation. Specifically, wtr+1 includes weight and bias matrices from WQ, WK, WV, WZ nodes in the Masked Self-attention sublayer and all linear transformations in the Position-wise Feedforward sublayer along with Final Feedforward Network.
  4. Model Aggregation: After deriving trained weights wtr+1 from all edge clients, the cloud server federates them and constructs a new global model version.
  5. Global Update: Finally, the cloud server broadcasts back the new configuration wt+1 to each edge device as to update the local models of Transformer-Fourier block. In the subsequent learning rounds, edge devices use the updated models to continue training and the communication process will repeat in order to optimize the local Transformer-Fourier models until the global learning model converges.

Essential devices for testbed in Federated-learning mode, the experiment is conducted using:

  1. CPU Intel Core i9-12900K
  3. Edge device: raspberry pi 4 (quantify: 05) 


Designing ECG Monitoring Healthcare System with Federated Transfer Learning and Explainable AI

Federated Transfer Learning and Explainable AI is developed and studied by the collaboration of ENSAIT, GEMTEX Laboratory – University of Lille, IAD – UDA, and School of Computing & Institute of Cyber Security for Society (iCSS), University of Kent, UK.


Figure 1.  Architecture of Federated Learning


Figure 2.  An overview of the proposed framework


Figure 3.  The architecture of the proposed denoising autoencoder


Figure 4. The architecture of the proposed denoising autoencoder



Figure 5. Overview of the proposed XAI module in our framework

Figure 6. Federated learning architecture in Healthcare system (i.e ECG)

Besides, we apply FL architecture in Heathcare (i.e. ECG) system. First phase, we develop methodology based on sample data. Then, we tend to collect data and extend our system by using IoT sensor (i.e. ECG) in person.

Figure 7. ECG signal displayed in IoT device


Location: Room 311B, Dong A University, 33 Xo Viet Nghe Tinh street, Hai Chau district, Da Nang city, Viet Nam


    Head of International Chair 

    Senior Associate Professor Kim Phuc


    MBA. Kim Duc Tran - Local

    MD. Tho Nguyen - Head of the international Unit of IAD, Coordinator &