This paper studies diagnosis-aid systems that use Artificial Intelligence tools. This kind of systems become very interesting in an uncertain industrial environment especially for flexible production systems. An overview of the most important artificial intelligence diagnosis tools is given. For each tool, we focus on diagnosis principles and on its advantages and disadvantages. That allows us to extract four important points that a diagnosis tool should fulfilled. Using these results, we propose a tool based on fuzzy Petri nets which allows to make a diagnosis using a model easy to build and that take into account the uncertainties of maintenance knowledges. This tool provides abductive approaches of fault propagations system with an efficient localization and a characterization of the fault origin. At the end, we apply our tool on an illustrative example of a flexible system diagnosis is presented.