Statistical Process Control - Course Work Example

The accompanying excel spreadsheet (Cells C2:C50) shows data on the number containers that break during the process of loading glass containers off the transport vehicles at a logistics base the data covers a 48-month period. The data suggests that, on average, 12.61% of the glass containers handled at the logistics base in a month end up in waste after breaking, with the standard deviation standing at 6.12%. Therefore, the process can keep wastage within at least 6.12% of the average breakage. The goal is to keep wastage within two standard deviations of the mean; this puts the control limits at 24.85% (upper control limit) and 0.37% (lower control limit). The following figure shows the process control chart.

Figure 1: Process Control Chart

The control chart shows that the process is not in control because some points are either at, or beyond, the upper control limit. However, no point is near the lower control limit, which suggests that, more often than not, breakages per month have been increasing. The process could benefit from Six Sigma tools because the lack of control suggests the need for improvement in quality management. Six Sigma tools help firms identify and eliminate the factors causing defects, which, in turn lowers process variability. For processes that are out of control, as the one in this case, implementing Six Sigma might require the review the goal (s) the goal in this case is to keep the product breakages within acceptable limits.

After the review of the process goal, it might be important to assess the factors influencing the rate of product breakage during materials handling. For instance, the operations manager can assess the layout of the logistics base to determine if it is constraining the movement of material handling equipment and personnel, thus increasing collisions that result in the breakage of the glass containers. Once the process manager has established the factors that have probably kept the process from staying in control, he/she can institute necessary changes and measure how the changes affect the process. In the preceding example, the process manager can change the layout of the logistics base to allow more space for the movement of equipment and personnel during material handling.

Following adjustments to the logistics base, the process manager should start collecting data on the percentage of glass containers broken during handling this data will provide a useful indicator of the impact of the facility layout on process outcomes. Analysis of the collected process data will help in the determination of the root cause of the deviation from the process control limits. It may seem plausible to argue that, if there are random instances of process outcomes falling beyond the control limits, there is insufficient justification for radical changes to a process. However, when numerous cases of process outcomes falling close to the control limit precede large, random deviations from the control limit, the random variations result from a series of minor, long-running deviations. If process managers do not address random variations from the control limit, the minor, long-running deviations from the mean eventually build up into significant variations from the control limit, which makes it difficult to attain process goals.