Towards predicting human error: Eye gaze analysis for identification of cognitive steps performed by control room operators

Today's plants use highly reliable equipment, state-of-the-art automation and control, and deploy sophisticated safety management regimes so as to make accidents rare. Despite these measures, a number of recent accidents point to the continued need to improve process safety. Various studies indicate that more than 70% of the accidents in process industries originate from human errors. Traditionally, human reliability is accounted for by using Human Error Probabilities, which emphasizes inevitability. In this work, we demonstrate that eye tracking serves as a reliable sensor of various cognitive tasks performed by the operator while managing process abnormalities. Experimental studies conducted on 11 participants reveal characteristic fixation patterns that contain information about their cognitive ability to orient, diagnose and execute recovery actions during abnormal situations. These results hold promise for predicting incipient cognitive failures which in turn would enable proactive measures to prevent human error.