Long term v. short term confusion
June 23, 2007 at 7:14 pm #47356
newbieParticipant@newbie Include @newbie in your post and this person will
be notified via email.
I am reading ´The Black Belt Memory Jogger´ from Six Sigma Academy, and have a question (pg 44 states):
Short term data is data collected from the process in subgroups. Each subgroup is collected over a short period of time to capture common cause variation only (ie not collected acorss different shifts because variation can exist from operator to operator)…For example, a process may use several raw material lots per shift. A representative short term sample may consist of CTQ measurements within one lot.
Again, I keep seeing this contradiction. The book speaks of the importance of representative samples (across all conditions), and then talks of working to avoid such things (avoiding multiple lots, operators, shifts, etc). How do you take short term samples that are so limited (within one shift, one operator, one lot, etc) and expect it to be a representative sample of the process? Is one supposed to take the short term sample, calcualte the short term PCI (eg Z score) and then account for ´representation´ using the dreaded 1.5 shift? Thanks!0June 23, 2007 at 9:07 pm #157867
Montgomery SPC 3rd editionParticipant@Montgomery-SPC-3rd-edition Include @Montgomery-SPC-3rd-edition in your post and this person will
be notified via email.
You are better off with following Montgomery’s SPC book. The Memory Jogger is a compilation of various generations of statisticians and can easily have some inconsistencies. Here are a few citations from Montgomery’s book including page numbers:
“subgroups or samples should be selected so that if assignable causes are present, the chance for differences betweent subgrops will be maximized whie the chance of differences due to these assignalbe causes within a subgroup will be minimized (143)”.
“Two general approaches to constructing rational subgroups are used. In the first approach, each sample consists of units that were produced at the same time (or as closely together as possible). Ideally, we would like to take consecutive units of production. This approach is used when the primary purpose of the control chart is to detect process shifts (144)”.
“In the second approach, each sample consists of units of product that are representative of all units that have been produced since the last sample was taken. Essentially, each subgroup is a random sample of all process output over the sampling interval. This method of subgroup sampling is often used when the control chart is employed to make decisions of acceptance of all units of product that have been produced since the last sample (145)”.
“The proper selection of samples requires careful consideration of the process, with the objective of obtaining as much useful information as possible from the control chart analysis (146)”.
Representative sampling in process control charting is also different from representative sampling in population parameter estimation. In process control charting your purpose is to determine assignable causes. As a result, you use all information available to develop your sampling plan. In that respect, it is non-random. In population parameter estimation the only way to get a “representative sample” is through random sampling.0
The forum ‘General’ is closed to new topics and replies.