SPSS On-Line Training Workshop
SPSS On-Line Training Workshop |
|
|
 In this Tutorial: |
|
||||
In this on-line workshop, you will find many movie clips. Each movie clip will demonstrate some specific usage of SPSS.
Quality Control: Techniques for process quality control are important for monitoring the quality of a process. Typically, there are two types of variability generated from a process. One is 'special causes'. These causes are due to unexpected causes that occur at a certain time in the production process. Statistical control charts such as x-bar and R-charts are used for monitoring special causes. In many incidents, these special causes are defectives, and actions should be taken to get rid of these causes. In some incidents, these causes may be a 'good' cause resulting into a better process. Actions should also be taken to retain such a 'good' cause for sustaining the quality.
Another type of variability generated from process is 'system variability'. This type of variability is usually generated by the process system itself. They are 'system causes'. These type of system causes are often more difficult to diagnose, since it is inherited in the system. It may take an extensive experiment in order to identify these causes. However, once the system causes that generate system variability is identified and resolved, the improvement is usually for a much longer term. SPSS provides tools for capability analysis for identifying the system causes, and three control charts for identifying special causes.
 |
Capability Analysis: The capability indices are often used to monitor the system variability. An incapable process requires the investigation of root causes that introduce the system variability. Such causes are called system causes. The following movie clip demonstrate how to conduct a capability analysis. |
The data used for this demonstration is the Ringdiameter(CasesareSubgroup) data set. See Data Set page for details. The diameter of piston ring is a quality measurement that need to be monitored in an engine manufacturing process. A random sample (aslo called a subgroup) of five pistons are selected each day for a total of 20 days. Each subgroup is recorded as a case. The five diameters are named as Diameter1 to Diameter5. The diameter is a continuous data, also call variable data.
|
X-bar & R-Charts: The X-bar and R-charts are used for variable data (continuous data) with the assumption that the data follows a normal distribution. X-bar monitors the process means, while the R-chart monitors the within group variation at a given time point. Another similar control chart for monitoring variable data is Xbar/s-charts. Both Xbar/R-charts and Xbar/s-cahrts are for subgroup sample size two or more. SPSS also provides the individual, Moving Range charts for situations where the subgroup sample size is one. The out of controls are highlighted in the charts for further diagnosis of special causes. |
The following movie clip demonstrates how to construct and apply the X-bar and R-charts to monitor a quality characteristic.
Click
here to watch how to construct and Interpret X-bar & R-Charts
The data set used for this demonstration is Ringdiameter data set. See the Data Set page for details. The diameter of piston ring is a quality measurement that needs to be monitored in an engine manufacturing process. Random samples of five pistons are selected each day for a total of 20 days. The diameter of the piston ring is measured. The diameter is a continuous data, which is also called variable data.
|
p- and np-Charts: In many processes, the quality characteristics are measured by the proportion of defective parts in a random sample of the product. p-chart is designed to monitor the proportion, while np-chart is for monitoring the number of defectives. |
The following movie clip demonstrates how to construct and interpret p- and np-charts.
Click
here to watch how to construct and interpret p- & np-charts
The data used for this demonstration is the proportion (or number) of defective cans of orange juice manufactured by a drinking company, the Orange Juice data set. See the Data Set page for details. The company examined the amount of orange juice in each can of a random sample of 50 cans per day. The defective is defined as the actual amount of orange juice that differ from the target 12 oz by .2 oz. The data used in this demonstration is the number of defectives in each sample for 30 days.
|
c- and u-Charts: Many quality characteristics are count of number of defects of a product. This type of count data often follows Poison distribution. c-chart is developed based the Poison distribution for monitoring the number of defects, and u-chart is often used to monitor the number of defects per unit of a product. |
The following movie clip demonstrates how to construct and interpret c- and u-charts.
Click
here to watch how to construct and interpret c- & u-charts
The data used for this demonstration is the number of defects per 400 square yards of cloth during the dyeing process in a cloth manufacturing company, the Dye Cloth data set. See the Data Set page for details. The data used in this demonstration is the observed number of defects of a roll of 400 yards for 20 randomly selected rolls.
This online SPSS Training Workshop is developed by Dr Carl Lee, Dr Felix Famoye , student assistants Barbara Shelden and Albert Brown , Department of Mathematics, Central Michigan University. All rights reserved.
