Discussion: Quality Control in Industry

QSO 510 Module Nine 1

Module Eight discussed contingency tables to summarize categorical data and a chi-square test for the independence of two categorical sets of data. Module Nine explores sources of process variation and the use of control charts to eliminate unusual sources of variation to ensure that an operations process is “on control.” There are two generally recognized aspects of quality: quality of design and quality of conformance (Montgomery, 1991). Quality of design refers to the type or grade of product that is to be made and its design. For example, a chemical company may produce two grades of product that are designed to be 95% and 99% purity, respectively. These streams of product are designed to be of different quality levels by virtue of differences in their purity levels. Quality of conformance refers to the uniformity of the product and how well it conforms to the specifications of the design. For example, a producer of electrical equipment that purchases a particular component from a supplier will want batches of the component to be within design specifications and as similar as possible. A more uniform product results in less scrap and rework by production employees. It also results in less inspection cost by the quality control department, lower long-run production costs, and increased productivity since the product can now be produced at a reduced cost. Greater uniformity of the product that leads to higher levels of quality of conformance is the purpose of control charting. The quality control chart is the primary tool in achieving greater levels of quality of conformance. Variation exists in all products and services. A production employee who takes readings on the flow pressure produced by an oil pump on an hourly basis will find that the readings are never the same. An employer who gives an entry exam to a prospective employee and then retests him the following day will obtain different results. A customer at a restaurant who is served by the same waiter for breakfast and then later for dinner will notice a difference in the level of service. Decreases in the variation of the level of a product or service result in improvements in quality of conformance. A control chart is a plot of observations from a process over time. Pegels (1995) suggests that "control charting involves the charting of statistics on a chart in such a way that deviations from a standard can be quickly observed, and action can be taken to correct the undesirable variation." Typically a process can be summarized by an average value (x bar) and a measure of variation, the range (R). The purpose of the control chart is process stability, through the reduction of process variability, which is done by distinguishing between common-cause and special-cause variation.

2 QSO 510 Module Nine

Common-cause variations are small, uncontrollable influences that are an inherent part of the process. They cannot be removed from the process without basic changes in the process that usually require management action. Special-cause variations are larger, unusual influences that can be removed from the process. A process that is operating with only common-cause variation present is said to be in-control. The control chart is used to determine whether a process is in-control. Control chart limits are typically set to a mean ±3 standard deviations from the mean. In general, when a process mean or variation exceeds control limits on a control chart, the process is said to be “out of control.” Control charts are widely used in industry and manufacturing. An operations manager may use a control chart to detect a broken pump that shows up as an out-of-control signal on the chart. A fast-food restaurant manager may detect lower productivity on a particular shift that falls below a lower control limit. A professor may use control charts to detect outstanding performers on a test that is revealed by an out-of-control signal.

QSO 510 Module Nine 3

References

Montgomery, D. C. (1991). Introduction to statistical quality control (2nd ed.). New York, NY: Wiley.

Pegels, C. C. (1995). Total quality management. Danvers, MA: Boyd and Fraser.