Affiliations: [a] Dofasco Inc., Hamilton, Ontario, Canada | [b] University at Buffalo, Department of Industrial Engineering, Buffalo, NY, USA
Correspondence:
[*]
Corresponding author: Colin G. Drury, University at Buffalo, State University of New York, Department of Industrial Engineering, 342 Bell Hall, Buffalo, NY 14260, USA. E-mail: [email protected]
Abstract: Three studies were performed to better understand the complex task of sheet steel inspection for multiple surface defects. Their aim was to combine techniques from off-line performance measurement studies with interviews and questionnaires coming from a more holistic tradition of inspection. In this way, more detailed insights were obtained from three studies of nine highly-trained and experienced steel inspectors. The performance study was designed to compare lighting conditions using a visual search task. It did not find lighting differences, but did find other task differences and the typical large individual differences. These individual differences were explored further with detailed interviews, a task analysis and a detailed questionnaire. Areas of significant agreement among inspectors were found, as were responses that differentiated high from low performing inspectors. Rasmussen's skill/ rule/ knowledge classification was used to aid the analysis of these latter studies, showing the contribution of higher-level cognitive processes to the inspection task.
