Key points are not available for this paper at this time.
The advent of Industry 4.0 and propelled the application of Artificial Intelligence in different industrial fields and contexts, such as predictive maintenance (PdM). Through its ability to assess the condition of equipment to detect signs of failure and anticipate them, PdM brings several potential benefits in terms of reliability, safety and maintenance costs among many other benefits. Different approaches are proposed in the literature. They are based on data, physic models or knowledge but several problems and limits persist, in particular, to override this dependence on a particular context, to utilize data and business knowledge considering the challenges of applying existing solutions to another context, difficulties associated with data analysis, and uncertainty management. In this context, the goal of this paper is also to highlight the challenges faced in the area of PdM, both for implementation and use-case. PdM remains a hot topic in the context of Industry 4.0 but with several challenges to be better investigated in the area of machine learning, knowledge representation and semantic reasoning applications.
Hafsi et al. (Sun,) studied this question.