The future ship bridge is foreseen to be an operations centre incorporating controls and monitors for all essential vessel function. This implies that the function of the human operator will be one of supervisory control: whereas the subsystems are automatically controlled, human supervision is needed to ensure that performance remains within specified ranges. In the present study we investigated how problem solving behaviour is affected by complexity and a priori probability of disturbances in a supervisory control task. An experimental set-up was designed in which subjects were required to supervise four independent shipping subsystems, navigation, propulsion, electricity and cargo, and to adjust the subsystems whenever deviations turned up. However, in order to apply the correct action, subjects first had to diagnose the cause of the deviation, by requesting information. Complexity and a priori probability were manipulated by varying the number of disturbances occurring simultaneously and disturbance rates over subsystems. Twenty four subjects participated in the experiment, all second year students attending a maritime institute. In general, the results suggest that subjects dealt with the disturbances sequentially, providing support for the "cognitive lockup" hypothesis: rather than dividing attention between problem solving and monitoring the rest of the subsystems, subjects focused on diagnosing the disturbances, ignoring the monitoring function. Across subjects, differences were found in reasoning abilities: some subjects could deduce more correct conclusions from the requested information in less time. Even though the results clearly indicate bottlenecks in information processing that could be reduced by decision support tools, we recommend to pursue the findings in a second experiment. In this experiment the effects of uncertainty and task goals on problem solving behaviour will be considered. Furthermore, the task will be more complex as subsystems will be interrelated. (Author/publisher)
Abstract