Artificial Intelligence Explainability (XAI) and Interpretability: Exploring the Potential of XAI in Fault Diagnosis and Cyber-Physical Systems

Dear Colleagues,

The application of AI has shown great potential in the past decade. Due to increasing computational resources and data, it is increasingly becoming possible to deploy AI in complex cyber-physical systems such as manufacturing systems, telecommunication systems, IOT-based systems to mention a few. Recent research completed by Deepmind, a leading AI company, has shown how machine learning, especially deep learning techniques, have great potential to reveal and inform how proteins, one of the basic building blocks of life, fold, called AlphaFold: a solution to a 50-year-old grand challenge in biology.

Nevertheless, models produced by current AI techniques, though powerful, are not easily understandable and remain a black box to most practitioners. As a result, it is not known how these models derive their conclusions and what lessons could be learned from their selected decision-making paths to deepen the understanding of the domain. Such an understanding could be important for ethical issues, transparency, and privacy concerns. When applied to fault diagnosis, it could support human engineers in diagnosing faults and providing transparency and explainability of why the faults occurred in cyber-physical systems. This would result in AI systems that can be totally trusted (trustworthy automated systems) and more accepted in manufacturing systems.

This Special Issue invites papers on the rapidly growing field of explainable AI (XAI) theories and methods, as applied to fault diagnosis of systems such as traditional manufacturing systems, smart cyber-physical systems, and remote condition monitoring of equipment to mention a few. We invite you to contribute to this issue by submitting both case studies and research articles, we are open to papers that address (but are not limited to) the following keywords:

• Application of fuzzy logic theory to aid understanding of AI decisions
• New explainable AI (XAI) concepts
• Novel cognitive architectures
• Natural language processing
• Equipment condition monitoring and maintenance
• Human in the loop systems
• XAI and Industry 4.0 / IoT
• Theories, analysis, and visualization of interpretable machine learning/deep learning method
• Industrial applications of interpretable machine learning/deep learning method
• Bayesian networks and probabilistic graphical models
• Knowledge representation and reasoning
• Reasoning under uncertainty

Dr. John Oyekan
Dr. Christopher Turner
Prof. Yuchun Xu
Dr. Ming Zhang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.