Statistical Process Control (SPC) uses statistical methods to track the variation of process outputs with the aim of maintaining those outputs within set limits. Unlike other forms of process control that are primarily concerned with controlling the elements of the process, SPC is primarily concerned with a consistent output of the entire process. As such, it is as much a management tool as it is a manufacturing tool—and an important contributor to product quality.
History
SPC originated in the early 1920s when Walter Shewhart developed the SPC chart, also called the Shewhart chart. SPC was used extensively during World War II to control the production of products important to the war effort, such as weapons and munitions. Its use declined after the war, but Japanese industry subsequently adopted it very successfully. Because of that success and the resulting competition with Western industry, particularly during the 1980s, Western companies began to vigorously re-employ SPC.
How SPC Works
Any process has at least one output, and each output has at least one attribute. For instance, a golf ball manufacturing process will have a golf ball as an output, and that golf ball will have several attributes: shape, elasticity, color, etc. SPC sets a target value for the attributes of a process, along with acceptable upper and lower limits. The variation of these attributes is monitored with the goal of keeping their values within the set limits.
Natural and Assignable Variability
The variability of attributes comes from two sources: assignable and natural. Assignable variations are those changes that can be “assigned” to specific causes. Natural variations are those statistical variations that are expected and predictable. An example of an assignable variation might be a shallow dimple in a golf ball that is caused by a mechanical fault of a machine. A natural variation might be small fluctuations in the gloss caused by differences in the raw materials used.
Control Charts
A typical SPC control chart will include a highlighted axis representing the target value with two other highlighted axes representing the upper and lower acceptable limits above and below it. If all the values fall between these limits, the variation is considered natural and the process is considered to be “in control.” If several points fall outside the limits, it is considered “out of control.” A quick glance at a properly designed SPC chart will tell a manager whether the process is in or out of control.
How SPC Helps
The excellent visibility of process consistency provided by SPC charts gives process managers a quick warning when things are not right within a particular process, and they often facilitate in diagnosing problems. Even if the causes of assignable variation are not easily discovered, the view of the process’s inconsistency makes resolution of the problem quicker and less costly in terms of manufacturing costs, customer satisfaction and product reputation.
History
SPC originated in the early 1920s when Walter Shewhart developed the SPC chart, also called the Shewhart chart. SPC was used extensively during World War II to control the production of products important to the war effort, such as weapons and munitions. Its use declined after the war, but Japanese industry subsequently adopted it very successfully. Because of that success and the resulting competition with Western industry, particularly during the 1980s, Western companies began to vigorously re-employ SPC.
How SPC Works
Any process has at least one output, and each output has at least one attribute. For instance, a golf ball manufacturing process will have a golf ball as an output, and that golf ball will have several attributes: shape, elasticity, color, etc. SPC sets a target value for the attributes of a process, along with acceptable upper and lower limits. The variation of these attributes is monitored with the goal of keeping their values within the set limits.
Natural and Assignable Variability
The variability of attributes comes from two sources: assignable and natural. Assignable variations are those changes that can be “assigned” to specific causes. Natural variations are those statistical variations that are expected and predictable. An example of an assignable variation might be a shallow dimple in a golf ball that is caused by a mechanical fault of a machine. A natural variation might be small fluctuations in the gloss caused by differences in the raw materials used.
Control Charts
A typical SPC control chart will include a highlighted axis representing the target value with two other highlighted axes representing the upper and lower acceptable limits above and below it. If all the values fall between these limits, the variation is considered natural and the process is considered to be “in control.” If several points fall outside the limits, it is considered “out of control.” A quick glance at a properly designed SPC chart will tell a manager whether the process is in or out of control.
How SPC Helps
The excellent visibility of process consistency provided by SPC charts gives process managers a quick warning when things are not right within a particular process, and they often facilitate in diagnosing problems. Even if the causes of assignable variation are not easily discovered, the view of the process’s inconsistency makes resolution of the problem quicker and less costly in terms of manufacturing costs, customer satisfaction and product reputation.