Abstract: A method and apparatus records downtime and the probable cause of the downtime of a network of programmable logic controllers and other microprocessor based devices for controlling of a manufacturing process. The method and apparatus also records the total downtime by separate causes and a total downtime of all causes encountered. The apparatus further records the number of times each cause occurred and a total number of all causes encountered. A graphical display provides a visual presentation of the number of times each cause occurred and the total downtime for each cause that occurred in Pareto chart form. The method provides for prioritizing alarms associated with the cause of a downtime and the apparatus has means for deciding which alarm notification, based on the alarm priority and therefore, its corresponding cause, recorded the most number of times it caused a downtime.