A method includes processing data associated with operation of equipment in an industrial process to repeatedly (i) identify one or more models that mathematically represent the operation of the equipment using training data and (ii) generate first indicators potentially identifying at least one specified condition of the equipment using evaluation data and the one or more models. The method also includes classifying the first indicators into multiple classes. The multiple classes include true positive indicators and false positive indicators. The true positive indicators identify that the equipment is suffering from the at least one specified condition. The false positive indicators identify that the equipment is not suffering from the at least one specified condition. The method further includes generating a notification indicating that the equipment is suffering from the at least one specified condition in response to one or more first indicators being classified into the class of true positive indicators.