Approaches to assessing the skills of operators of complex technical systems
Abstract
Presented techniques are developed to support instructor assessment process for operators of complex technical systems and make it possible to: support the express outcome grading for current operator’s activity by means of comparing their parameters with the activity patterns collected beforehand in the relevant database; carry out quantitative and qualitative assessments of operator’s skill class, which are based on comparisons of current operator’s activity with the database patterns in the wavelet representation metric associated with observed parameters, with instructor’s assessment/supervision comments being employed, as well as probabilistic assessments of skill class recognition using both sample distribution functions of distances to cluster centers in a scaling space and Bayesian likelihood estimations; optimize training process by means of selecting next actions that are useful from the viewpoint of operator preparation. An implementation of the approach in question should be useful as a tool for comparing different training means and syllabus and, in perspective, for replacing “a subjective instructor” by “the standardized instructor” at the operator’s training process.
