Condition monitoring is a lively research domain, with hundreds of researchers sharing new techniques coping with complex problems, such as variability of the operating conditions, complex machinery monitoring, diagnostics, prognosis... Considering the many methods that have been proposed these last decades, it is striking how few of them have reached the industrial application. One of the reasons is that the intensive validation of these methods over real-life data required to ensure their reliability is often difficult to achieve. The results obtained over one dataset may not be easily reproduced over another one. Furthermore, can the comparison between two techniques be objective, when applied to different datasets?

Some laboratories and companies generously share the datasets recorded on their test benches or industrial devices. Unhappily, it is often difficult for them to know for certain who has worked on their data, which is both frustrating and a real problem when they report to their authority about the diffusion of the data, in order to get the subsidies required for the test bench exploitation. From the user's point of view, there can also be difficulties in finding the right datasets for a specific study among the jungle of all that is proposed on the web. How to find the appropriate dataset for a prognosis study, or a fatigue study, or the study of any specific kind of damage? Another problem can be met by those who wish to share their datasets with the scientific community but do not have the technical skills or staff to do so.

CM Base is a web portal that aims at facilitating the sharing of the data, by offering many a functionality, such as a list of all the existing datasets and test benches with all related papers, searching facility allowing to extract from the base all datasets related to a specific problem or papers related to a specific dataset.