Automating human skills : preliminary development of a human factors methodology to capture tacit cognitive skills

Abstract

Despite technological advances in intelligent automation, it remains difficult for engineers to discern which manual tasks, or task components, would be most suitable for transfer to automated alternatives. This research aimed to develop an accurate methodology for the measurement of both observable and unobservable physical and cognitive activities used in manual tasks for the capture of tacit skill. Experienced operators were observed and interviewed in detail, following which, hierarchical task analysis and task decomposition methods were used to systematically explore and classify the qualitative data. Results showed that a task analysis / decomposition methodology identified different types of skill (e.g. procedural or declarative) and knowledge (explicit or tacit) indicating this methodology could be used for further human skill capture studies. The benefit of this research will be to provide a methodology to capture human skill so that complex manual tasks can be more efficiently transferred into automated processes.