An in-depth analysis of maintenance-related events was performed by screening four different databases. The events cover the period 2002 to 2013. A total of 921 events were selected for analysis. An examination of the selected events resulted in their classification into nine categories or groups (e.g., plant state, type of maintenance, affected component, root cause, etc.). For further analyses, the categories were divided into families and, if necessary, into subfamilies. One of the event classifications was according to the type of maintenance (periodic, predictive, planned, and corrective). The data analysis indicated that 47% of the events reported were related to periodic maintenance. The main affected components were valves (with 33% of the events), followed by electric power components (23%). The main root causes observed are maintenance performed incorrectly (27%), deficiencies in written procedures or documents (19%), and deficiencies in management or organization (17%). Regarding the impact on safety, the dominant family is potential effects on safety function (57%), followed by significant effect on operation (20%).