The objective of this study is to investigate experimentally the relationship between an operator’s mental workload and the information flow rate of accident diagnosis tasks and further to propose the information flow rate as an analytic method for measuring the mental workload.
There are two types of mental workload in the advanced MCR of NPPs: the information processing workload, which is the processing that the human operator
must actually perform in order to complete the diagnosis task, and emotional stress workload experienced by the operator. In this study, the focus is on the former. Three kinds of methods are used to measure the operator’s workload: information flow rate, subjective methods, and physiological measures.
Information flows for eight accident diagnosis tasks are modeled qualitatively using a stage model and are quantified using Conant’s model. The eight accident cases are considered here are: Loss Of Coolant Accident (LOCA), Steam Generator Tube Rupture (SGTR), Steam Line Break (SLB), Feedwater Line Break (FLB), Pressurizer (PZR) spray and heater failure, Reactor Coolant Pump (RCP) trip, Main Steam Isolation Valve (MSIV) failure, and PZR spray failure. The information flow rate is obtained for each diagnosis task by imposing time limit restrictions for the tasks.
Subjective methods require the operators to respond to questionnaires to rate their level of mental effort. NASA-TLX and MCH scale are selected as subjective methods. NASA-TLX is a subjective method used in the various fields including the aviation, automobile, and nuclear industries. It has a multi-dimensional rating technique and provides an overall workload score based on a weighted average on six subscales using pair-wise comparison tests. MCH, on the other hand, is one-dimensional and uses a 10- point rating technique. As with NASA-TLX, the higher the score is, the higher the subjective workload is. For the physiological measurements, an eye tracking system analyzes eye moveme...