This paper describes a new technique based on rough sets to extract decision rules from large volumes of data captured by protection, control, and monitoring intelligent electronic devices. The methodology correctly identifies faults from large datasets and could be used to assist operators in their decision-making processes. Building knowledge for a fault diagnostic system is a time-consuming and costly process. The quality of a knowledge base can sometimes be hampered by a large number of superfluous decision-making rules that can lead to an unnecessarily large knowledge base system and inefficient or even detrimental rule maintenance. The methodology proposed cannot only induce decision rules efficiently but can also reduce the size of the knowledge base without causing loss of useful information. Results can be used by an expert system to generate supervisory automation and to support operators, for example, during an emergency situation. This methodology involves the generation of human-machine interface alarms. These can then be used for diagnosis of the type and cause of a fault event to give suggestions for network restoration and post-emergency repair. A power systems computer aided design/electromagnetic transients including dc simulator has been used to investigate the effect of faults and switching actions on the protection and control equipment associated with a typical distribution network. The fundamental ideas of rough set theory are discussed, followed by a rule assessment method that is outlined using an illustrative example.