In the eighth chapter of their book Six Sigma: The Breakthrough Strategy Revolutionizing the World’s Top Corporations, the authors Mikel Harry, Ph.D., and Richard Schroeder discuss “Measuring Performance on the Sigma Scale.” In the previous chapter, the authors discussed the Breakthrough Strategy of implementing Six Sigma on the business, operational, and process level.
In this chapter they focus on the question “how does improving the Sigma level of a company’s processes improve that company’s performance?” One of the ways this is done is by setting a control limit in order to control the variation within the specification limit. However, even when one thinks one has controlled the variation to a certain Sigma level, it turns out that the long-term Sigma level of the process ends up, on average, 1.5 Sigma less than the short-term improvement one thought one had achieved. Why is this? Because of something called “process drift”.
The very illuminating example used by the authors is that of designing a garage to accommodate a vehicle’s width. Let’s assume that you had an architect who was going to design your garage to accommodate your vehicle, and that you only have one vehicle to park in that garage. What the architect has to accommodate is not just the width of the vehicle, but the variations in which it will be driven into the garage. Yes, the garage needs to be as wide as the vehicle, but what if the driver is coming into the garage slightly off center? Just how much is the variation between individual drives, not to mention individual drivers in the household? Also, there needs to be some accommodation not just for the width of the car, but the width of the driver.
When the driver gets out of the vehicle, there has to be enough room for the driver to be able to fit between the vehicle and the garage wall as the driver makes his or her way towards the door leading into the house. Depending on how much is stored in the sides of the garage, the garage will need to accommodate the width of the storage as well.
The “process drift” is analogous to the variation in an individual’s centering of the vehicle from day to day. This can be effected by a) the amount of sleep the person has received the night before, b) the amount of alcohol the person has had before driving home, and c) the amount of light outside the garage at the time the driver approaches it.
The analogy in the world of manufacturing that causes “process drift” can come from three primary sources or variation:
1) Inadequate design margins
These need to account for natural variation. In the case of the garage, the architect needs to account for the fact that the car will not be perfectly centered, and so a reasonable amount of variation will be needed to be designed in.
2) Unstable parts and materials provided by vendors and suppliers
Vendors and suppliers will always be on the lookout to change parts and materials to something cheaper. This will sometimes cause variation that the manufacturer needs to be on the lookout for.
3) Insufficient process capability
This means that the process is not capability of meeting the specifications limits of the critical-to-quality characteristics that customers demand. If an engineer does not take the width of the driver into account as well as the width of the automobile, the driver may complain that they can get the car into the garage, but are subsequently unable to get out of the car, which understandably would prove irksome to the driver.
These three sources of variations can occur individually or, more often than not, can overlap and happen all at once. Six Sigma is used to tease out these three common sources of variation, and thus to help remove “process drift”.
In the next post, the authors go into a little more detail regarding the relationship between a customer’s critical-to-quality characteristics (CTQs) and the specification limits a manufacturer sets in order to make sure that they are operationally satisfied in the manufacturing process.
Filed under: Uncategorized |
4squareviews 
Leave a Reply