Identifying variability key characteristics for automation design - a case study of finishing process

This paper describes an investigation of human interaction with process variability (i.e. variability not introduced by the humans themselves) in a manual manufacturing process. The process studied is grinding-polishing of high-value metal components, to evaluate the extent of the variability and how the operators applied their skills to overcome it. The research methods include analysis of documentation, observation and video recording and interviews. The results indicate that humans are able to adapt to variability in the parts and tools in order to deliver the product within specification. This suggests unconscious and automated behaviour meaning that the procedures executed are embedded in the minds of the operators. Vision and tactile senses were mainly used to check work progress and control critical features (Key Characteristics). Based on the findings of this and other case study, a framework will be developed to categorise variability in manual manufacturing processes to support the design of an automated solution.