The effects of three statistical control charts on task performance

This study investigated the effects of graphical characteristics on three common statistical process control (SPC) charts, Shewhart, exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) charts, with a near-equivalent statistical performance. Three main tasks, identification of out-of-control (OOC) points, estimation of process mean and identification of process pattern, were designed to evaluate the effectiveness of these charts. Participant response time and accuracy were collected. Results showed that each SPC chart performed equally well in perception and identification of OOC. For the mean shift task, charts yielded the fastest response times, and both and EWMA charts produced similar accuracy. Hoever, all SPC charts led to consistently poor performance in the pattern tasks. For subjective preferences, participants reported a preference for charts in all experimental tasks, and the only difference noted between the EWMA and CUSUM charts was for the pattern task. These findings suggest the advantages of charts, which may account for their widespread use in industries, over EWMA and CUSUM charts. Computationally, charts were easier to understand and, graphically, they provided more direct information. Thus statistical performance as well as human performance using these tools are important elements to consider in the selection of the appropriate SPC technique.