One of the more interesting aspects of SPC is the successful implementation of control charts in a job-shop manufacturing environment. Most job-shops are characterized by short production runs, and...

One of the more interesting aspects of SPC is the successful implementation of control charts in a job-shop manufacturing environment. Most job-shops are characterized by short production runs, and many of these shops produce parts on production runs of fewer than 50 units. This situation can make the routine use of control charts appear to be somewhat of a challenge, as not enough units are produced in any one batch to establish the control limits. This problem can usually be easily solved. Since statistical process-control methods are most frequently applied to a characteristic of a product, we can extend SPC to the job-shop environment by focusing on the process characteristic in each unit of product. To illustrate, consider a drilling operation in a job-shop. The operator drills holes of various sizes in each part passing through the machine center. Some parts require one hole, and others several holes of different sizes. It is almost impossible to construct an and R chart on hole diameter, since each part is potentially different. The correct approach is to focus on the characteristic of interest in the process. In this case, the manufacturer is interested in drilling holes that have the correct diameter, and therefore wants to reduce the variability in hole diameter as much as possible. This may be accomplished by control charting the deviation of the actual hole diameter from the nominal diameter. Depending on the process production rate and the mix of parts produced, either a control chart for individuals with a moving range control chart or a conventional and R chart can be used. In these applications, it is usually important to mark the start of each lot or to batch carefully on the control chart, so that if changing the size, position, or number of holes drilled on each part affects the process the resulting pattern on the control charts will be easy to interpret.