Data Collection

Proper data collection is very important to determine where we are and where we want to go. From the collected data we know our process behavior, the entitlement of our process and many more information can be assessed.  The precision, consistency, and strategies employed during the collection of data have a large impact on our ability to accurately analyze, improve, and control.

The Data Collection Sheet is a form used to capture and organize measurements. The data collected using Data Collection Sheets and Control Cards is then entered into a database for further analysis.

The results of sampling are greatly influenced by the method of selecting the sample.

Randomization of sample selection increases the probability that the sample is representative of the population. A random sampling is one where the conditions are such that each unit of the population has an equal chance of being chosen.

Stratified sampling is used when the lots are the combination of differences such as:  shifts, machines, employees, etc.

In sequential sampling, samples are selected in the order in which they are produced.

The best technique of data sampling will be based on the process behavior itself. Thus It is important for a black belt or green belt to explain several method of sampling and technique to the team members ( subject matter expert ) and let’s everybody agreed with the data collection approach to undertake.

Some of the consideration for sampling strategies will be things like production mode, the batch production schedule and the timing of the lots. Upon the agreement, the use of data collection sheet is vital in the process of data taking. Below is some of the example from SSA for data collection sheet. You may have your own data collection sheet.

Let’s have a look on some sampling method graphically which I take from my reading.

Random Sampling

Samples need to be selected at random to increase the likelihood that the measurements of the samples are representative of the process population. It improves the validity of population estimates and statistical tests such as capability analysis or the calculation of the probability of defect. Sampling only the first few parts does not show how the process performs over time.

To conduct Random Sampling requires that random numbers be generated and that the numbers can be applied to the process output. Random numbers can be generated using random numbers of tables, Excel or Minitab. In Minitab use, Calc > Random Data > Uniform.

In this example the the team member will collect data from sample 3 , 22 , 62 , 16 , 37 , 71 etc.

Sequential Sampling

In this example Sequential Sampling simply involves collecting the sample of five bolts each hour.

A sequential sampling can also be applied to transactional processes. One example would be a survey in a cafeteria. We could select the first five people arriving at 7:00, 12:00, and 6:00.

Stratified Sampling

In this example, the lot is the result of multiple operators and machines, the product is actually the combination of a several smaller lots. Therefore, we want to use a Stratified Sampling plan.

For example from smaller lot  I , sample 37 , 22 , 62 ,16, 71 and 3 the same method goes to smaller lot II and III.