CPMXai:Cognitive Predictive Maintenance and Quality Assurance using Explainable Ai and Machine Learning

The practice of predictive maintenance has escalated since the advancement in Artificial Intelligence (AI) and Machine Learning (ML). It anticipates the maintenance required, avoiding unnecessary costs (saving time, energy, money and resources) and breakdowns of machines. However, for more accurate and better predictions cognitive predictive maintenance is required. The AI/ML for cognitive predictive models require all algorithms to be based on supervised and unsupervised learning, requires labelled data where the amount of data is huge as it comprises historical data, sensor data, related proprietary resources and many more. Again, the decisions generated by the model can also be difficult to comprehend without any explanation. CPMXai aims to resolve these issues by forming a collaboration between the leading industry partners, SMEs, research institutes and universities. The collaborated consortium comprises expert personals from the different entities with experience, skills and knowledge to these problems. CPMXai has 3 objectives i.e., 1) identify use cases in the industries, 2) develop a new automatic data labelling tool with the help of digital twin and lastly, 3) develop a self-monitoring, self-learning, self-explainable system to predict. CPMXai will develop a digital twin for cognitive predictive maintenance through automatic data labelling, AI/ML and Explainable AI (XAI) to reduce unwanted situations and enhance maintenance in manufacturing and production processes. This will later be generalized and applied in other industries meeting their requirements and resulting in sustainable manufacturing and increasing the competitiveness of Swedish
manufacturing.

AI system with Explanation for industry

[Show all publications]

Attention-Based Fuzzy Neural Networks for Self-Supervised Data Annotation (Jan 2026)
Md Rakibul Islam , Shahina Begum, Mobyen Uddin Ahmed, Shaibal Barua
Intelligent Systems with Applications (ISWA)

Integration of Explainable Artificial Intelligence and Multimodal Machine Learning for Drivers’ Fitness (Oct 2025)
Mobyen Uddin Ahmed, Arnab Barua, Mir Riyanul Islam, Shaibal Barua, Shahina Begum
International conference on Advanced Machine Learning and Data Science (AMLDS 25)

A Theoretical Probabilistic Framework for Explaining Generative AI (Oct 2025)
Shahina Begum, Shaibal Barua, Mobyen Uddin Ahmed, Mir Riyanul Islam
International conference on Advanced Machine Learning and Data Science (AMLDS 25)

Privacy-preserving Ground-truth Data for Evaluating Additive Feature Attribution in Regression Models with Additive CBR and CQV (Oct 2025)
Mir Riyanul Islam, Rosina O Weber , Mobyen Uddin Ahmed, Shahina Begum
Knowledge-Based Systems (KBS)

Interpretable Self-Supervised Learning for Fault Identification in Printed Circuit Board Assembly Testing (Sep 2025)
Md Rakibul Islam , Shahina Begum, Mobyen Uddin Ahmed
AI-Based Machinery Health Monitoring (ApplSci25)

Towards Sustainable AI Development: A Focus on Transparency and Explainability (Jul 2025)
Shahina Begum, Mobyen Uddin Ahmed, Mosarrat Farhana
2nd International Conference on Creativity, Technology, and Sustainability (CCTS)