Abstract

The use of data-driven techniques by the nuclear technology industry to enhance asset availability, safety, and dependability has expanded. To build and implement such systems successfully, it is crucial to comprehend the foundational concepts between the disciplines. This study examines the foundations of artificial intelligence and the current state of learning-based approaches in nuclear science and engineering in order to assess the benefits and drawbacks of using such techniques for nuclear applications. This research focuses on applications in three significant decision-making and safety-related subfields. These include radiation detection, reactor health and monitoring, and optimization. Recent studies are examined, and the fundamentals of learning-based methodologies used in these applications are discussed. Additionally, as these techniques have improved in use throughout. Additionally, it is anticipated that learning-based methods will become more popular in nuclear science and technology as they have become more useful over the past ten years. As a result, it is important to understand the advantages and challenges of using such methodologies in order to improve research plans and recognise project risks and opportunities.