Issue link: https://iconnect007.uberflip.com/i/597087
November 2015 • The PCB Magazine 27 should be controlled. They can also be instru- mental in determining the appropriate process specification limits; for example, a chemical supplier should know the optimum operating parameters for the product they are supplying. Again, and I can't say this enough, customers want to see the processes controlled that you have deemed critical to your business; no one else can made that determination for you. Variation The one constant about any process is that there is inherent variation within it; in other words, variation is a given. No two things are exactly alike, and our personal experience proves this out on a daily basis. We often use the expression "Like two peas in a pod" to de- scribe two things that, on the surface, appear to be the same. But if we took this literally and opened the pod, close examination would find differences in the size, shape, color, texture and blemishes of the two peas. Now the critical as- pect of this is "Does this variation really mat- ter?" If the only requirement is tasty peas, than this variation absolutely does not matter. Although no two things are exactly alike, when talking about a process the goal should be to reduce the amount of variation between parts as much as possible. To do this, it is impor- tant to understand the difference between the two causes of variation; common and special. Common Cause: Also called random or normal, this variation is inherent in the process, affects every part, is repeatable, and most im- portantly, is predictable. Common cause varia- tion is created by many factors that are part of the process and are acting totally at random and independent of each other. Their origin can usually be traced to the key elements of the system within which the process operates (ma- terials, equipment, people, environment, meth- ods). If only common causes of variations are present, the output of the process forms a distri- bution that is stable over time. An example of common cause variation would be the amount of time it takes you to drive to work. Taking the same route every day, there would be variations in time that are due to which stop lights you make or miss, minor traffic fluctuations, and how closely you follow the speed limit. Special Cause: Also called assignable, these causes of variation are not a normal part of the process, does not affect every part, and can be identified to a specific, special cause. It means that something about the process has changed and has been created by a non-random event leading to an unexpected change in the process output. The effects are intermittent and unpre- dictable, and thus dangerous. If special causes of variation are present, the process output is not stable over time and is not predictable. Us- ing the above example, a special cause would be a weather-related delay on the drive to work. Control vs. Capability The word "control" often has a negative connotation surrounding it, which in some us- ages is well deserved. However, from a statistical standpoint, control refers to the consistency of a process; the ability to produce repeatable prod- uct within certain predetermined limits. As we learned above, all processes have variability, and a process is deemed to be "in statistical control" when it is affected only by common cause varia- tion. In contrast, capability refers to the ability of a process to produce output that meets speci- fications, in other words, shippable parts. There are various levels of capability which we will dis- cuss a bit later, but in general a process is said to be capable if nearly 100% of the output from the process is within the specifications. The goal of statistical process control is to develop a process that is both in control and capable. A key point that may often be hard to un- derstand is that a process can be statistically in control but not capable. This means that the normal variation in the process will not allow FeATure DATA AnALyTICS THroUGH STATISTICAL TeCHnIQUeS " The word "control" often has a negative connotation surrounding it, which in some usages is well deserved. "