Plant operator´s activity model and accident cause analysis based on mental images

Operator´s activities in a plant operation are basically regulated by their images with regard to the status quo (mental images of the system state) and the target state of the system (mental images of the ideal system state). The paper tries to describe operator´s actions using the framework of a “prediction model”, where an operator finds a discrepancy between his/her mental images and then takes an action depending on various kinds of information available to him/her. Both real time and cumulative information are categorized and the relation between information and cognitive process in the operator´s mind is reviewed. Specialists of accident cause analysis are responsible for finding out not only the direct cause but also latent ones behind the scene using surface evidences of an accident. The proposed framework will help their work