Abstract

Over the last eight years an extensive program on the modelling of human supervisory control behavior and performance has been carried out at the Delft University of Technology. This paper will summarize the positive and negative results as they have been experienced over this period.Modelling human supervisory control behavior is not just an interesting scientific activity, it is meant to result into engineering tools on the basis of which a control room can be designed optimally in terms of adapting it to man's capabilities and avoiding his limitations. In fact, one aims for a quantitative description or model where the influence of task variables, such as display configuration or dynamics of the system to be supervised, can be predicted on the human supervisory control behavior. Such models exist for manual control, but for supervisory control most of the modelling attempts have failed. This fact is not only caused by the variability in supervision tasks. Certainly, there is a great difference in supervisory tasks during normal and abnormal operation of a plant. In normal operation the major task is one of optimizing the plant e.g. by controlling the setpoints, whereas in abnormal operation, tasks like fault management and intervention have to be executed. Moreover, during start-stop procedures a mixture of these tasks occurs. But, the fact that in a particular process control condition, tasks are poorly defined in a quantitative way, is probably the most important reason for the many disappointing results. In fact, the following statement can be made:Only for those tasks which are well defined, the modelling of human supervisory behavior and performance may be successful.In this paper our experience on human supervisory control behavior modelling in different experimental setups will be discussed. It will be emphasized that the approach as followed in manual control studies cannot be followed here; directions for further research will be given

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call