Abstract
In almost every process control situation the human supervisor has to attend many sources of information. With large numbers of information sources the supervisor is forced to divide his attention and to acquire information through sampling. In the 60′s and 70′s several models of sampling with different criteria for optimal behaviour, have been proposed. Behavioural data indicate, however, that human monitoring behaviour can show large deviations from optimality. To account for these deviations a new model based on sequential sampling, is proposed. Experimental testing of this model reveals the development of fixed sampling sequences and the choice of a time base for execution as important performance shaping factors.
Get full access to this article
View all access options for this article.
