Limits to human optimization in inspection performance

As inspection moves from unaided human skills to human-computer hybrid task s, there is a need for models of the human and the computer which have comm on parameters. With appropriate models, functions can be allocated to produ ce optimal designs, and assistance provided to the human inspector via job aids and training. A model was developed of the human in a two-component co mpound inspection task consisting of search and decision. Optimizing this m odel showed that the choice of optimal values of parameters in the two subm odels was independent. Ten subjects were tested on a two-component inspecti on task, using components which had earlier been validated separately. Subj ects showed some aspects of optimum behaviour, for example sub-model indepe ndence, stopping the search after an integral number of scans, and varying their decision criteria to respond to the probability and cost structure. H owever, in this more complex task, subjects often reverted to simpler decis ion rules, for example always stopping the search after one scan or accepti ng (or rejecting) all potential defects detected. The implication for hybri d automation systems is that humans will need help such as job aids or trai ning if they are to perform optimally when given both search and decision t asks.