In this chapter we studied how to improve business processes by using control charts. We began by considering several meanings of quality, and we discussed the history of the quality movement in the United States . We saw that Walter Shewhart introduced statistical quality control while working at Bell Telephone Laboratories during the 1920s and 30s, and we also saw that W. Edwards Deming taught the Japanese how to use statistical methods to improve product quality following World War II. When the quality of Japanese products surpassed that of American-made goods, and when, as a result, U.S. manufacturers lost substantial shares of their markets, Dr. Deming consulted and lectured extensively in the United States . This sparked an American reemphasis on quality that continues to this day. We also briefly presented Deming's 14 Points, a set of management principles that, if followed, Deming believed would enable a company to improve quality and productivity, reduce costs, and gain competitive advantage.
We next learned that processes are influenced by common cause variation (inherent variation) and by assignable cause variation (unusual variation), and we saw that a control chart signals when assignable causes exist. Then we discussed how to sample a process. In particular, we explained that effective control charting requires rational subgrouping. Such subgroups minimize the chances that important process variations will occur within subgroups, and they maximize the chances that such variations will occur between subgroups.
Next we studied 
While it is important to bring a process into statistical control, we learned that it is also necessary to meet the customer's requirements (or specifications). Since statistical control does not guarantee that the process output meets specifications, we must carry out a capability study after the process has been brought into control. We studied how this is done by computing natural tolerance limits, which are limits that contain almost all the individual process measurements. We saw that, if the natural tolerance limits are inside the specification limits, then the process is capable of meeting the specifications. We also saw that we can measure how capable a process is by using sigma level capability, and we learned that a number of major businesses now orient their management philosophy around the concept of six sigma capability. In particular, we learned that, if a process is in statistical control and if the process has six sigma or better capability, then the defective rate will be very low (3.4 per million or less).
We continued by studying p charts, which are charts for fraction nonconforming. Such charts are useful when it is not possible (or when it is very expensive) to measure the quality characteristic of interest.
We concluded this chapter with an optional section on how to construct cause-and-effect diagrams and defect concentration diagrams. These diagrams are used to identify opportunities for process improvement and to discover sources of process variation.
It should be noted that two useful types of control charts not discussed in this chapter are individuals charts and c charts. These charts are discussed in Appendix L of the CD-ROM included with this book.