Question

The first step in selecting a Shewhart chart is to determine the type of data you are collecting. For this discussion, identify and discuss the 2 types of data that can be used with this type of chart. What are some examples of each type? Be sure to discuss the subdivision of each type. Based upon your response to these questions, why is this step so important? What would happen if you skipped it?

Answer 1

Schewart chart can be constructed from different kinds of data, viz. attribute or continuous ones, following different type of distributions like Binomial, Poission or Gamma.

One of the common types of data is attribute data, which may follow Poisson or Binomial distribution. This includes count, which is used to construct c chart, where the data is in from of the number of faults from a given sample. Another form of data is as proportion of defective items from the given sample size, used in constructing p chart. Other type of data comes through a continuous process, such as weight of produced items, which can be used to construct Xbar and R charts.

The attribute data can be classified into two categories viz. Binomial data which takes into account two alternatives ( pass /fail or defective / Non defective) for classifying items within data. P chart where proportion of defective items in a sample is an example of binomial data. Other type is poisson data which is in from of counts, such as number of defects in a sample ( C chart)

Th continuous data can be univariate or multivariate depending on the number of variables to be measured during study. Examples of univariate data is study of process temperature taken every 12 hours throughout the week. A multivariate data can have the readings of temperature, pressure and humidity of the process environment at the same time.

Subdivision is necessary to identify the data characterstics and identification of the most suitable method of analysis ( control chart) for effective analysis of data. If the division is not done correctly, the results may not communicate the real underlying causes which need to be identified through the study.