Skip to Content
Overview

Statistical Process Control is the process of detecting systematic deviations from the targets maintained for a particular quality characteristic. The central tool to carry out this analysis is the Control Chart.

Control Chart

The control chart is a graphical tool that tracks one or more control variables of the characteristic. The most important variables are:  a) Mean value , b) Standard deviation  S, c) Median value, d)Range, e)Original value of a sample, f)Number of nonconforming units, g)Number of defects. The chart also contains action limits and/or warning limits. These action limits can be used to trigger the valuation of the inspection characteristic result as also for triggering workflow if defect recording for characteristic is set up.

The control chart is different from the run chart. Whereas the run chart also plots the individual measured values or mean values, it does not contain action limits. Instead it contains the tolerance range limits.

SPC is generally used to monitor the production process and correct necessary processes if consistent deviations are observed. However, it can also be used in monitoring the quality of materials received from a vendor, or the quality of materials shipped to/returned from customers.

The most commonly used control charts are Acceptance charts and Shewart charts. Others include, a) Moving mean-value chart, b)EWMA chart, c)Original value and moving range chart for sample size n = 1, d) NP- chart for the number of nonconforming units, e) P-chart for the fraction of nonconforming units, f)C-chart for the number of defects, and g) U-chart for the number of defects for each sample unit. The appropriate chart is maintained in the sampling procedure.

Displaying/Changing Control Charts

Control charts can be called either from the “Results Recording” screen or from the transactions QGC1 (Control charts for inspection lots), QGC2 (Control charts for task list characteristics), and QGC3 (Control charts for master inspection characteristics). Each control chart is given a chart number by the system. This number is generated the very first time results are recorded against an inspection lot (provided SPC settings have been maintained). The status of these control charts is “New.” Activating the chart changes the status to “Active.” When not needed anymore, the chart can be closed, and the status changes to “closed.” If the SPC criterion is maintained for a given material/characteristic across several lots, then a new number is not given unless the old chart is closed.

 Calculation of Capability Indexes

There are two important measures that are used for measuring the process capability: Cp and Cpk values.

 For a quantitative characteristic, the values are calculated as follows.

Cp = (USL – LSL) / 6*s

Cpk = min((USL – ), (  – LSL)) / 3*s

Where USL and LSL are the tolerance range limits and  and s are the estimated values of the mean and the standard deviation of the distribution.

For a qualitative characteristic the Cpk value is calculated as Cpk = u(1-p) / 3 where p is the estimated share of nonconforming units and u is the quintile function of the normal distribution.

Setting up SPC in SAP

There are three important steps in setting up the correct SPC parameters. 1) Maintaining correct settings in the sampling procedure, 2) Maintaining correct settings in the Master Inspection Characteristic (MIC), 3) Maintaining desired SPC criterion in the task list.

  1. Maintaining Sampling Procedure: Maintaining Sampling Procedure is a must for implementing SPC. The following data must be maintained properly in the Sampling Procedure.
    1. Sampling Type: Only “Fixed sample” is allowed.
    2. Valuation Mode: If valuation is desired based on action limits, “SPC Inspection” valuation mode must be selected. However, this does not work with moving mean-average charts, EWMA chart, and moving range charts.
    3. Control Chart Type: The appropriate control chart type needs to be selected.
    4. Sample size: If both internal dispersion and process dispersion data are required, sample size has to be greater than “1.” By definition, for sample size n = 1, the internal dispersion is always zero.
  2. Maintaining MICs: The following parameters must be maintained (as appropriate for different types of charts).
    1. Measured values
    2. Upper and Lower specification limits
    3. Confirmation of number of defects
    4. Single result
    5. Scope not fixed/ Fixed Scope/Smaller scope/Larger scope
    6. If valuation based on violation of action limits is desired, then indicator for defects recording must be set, and appropriate defect codes maintained.
  3. Maintaining criterion in task list: The SPC criterion decides whether a separate control chart will be created for each inspection lot, or for each Material/Characteristic combination across several lots, or for each Material/Characteristic/Vendor combination, etc. There are several standard criteria available in the system, but new ones can be defined in customizing.

 Calculation of control limits

The calculation of control limits is maintained in customizing. Up to five control limits can be set up for each track (control variable) on a control chart: a) Upper action limit, b) Upper warning limit, c) Center line, d) Lower warning limit, and e) Lower action limit. The algorithms for the calculation of these limits are based on standard statistical methods and are already maintained in the system. However, if it is desired that users be able to manually change the control limits, the indicator “Change control limits” must be maintained in Customizing.

Setting parameters for reference axis

The parameters for the reference axis can be set in customizing. The options available are, a) Object number (inspection lot, sample, inspection point), b) Creation time (of data record), c) Time of last change (to data record), and d) Inspection time.

 SUMMARY

Statistical process control benefits organizations by providing a systematic method for the monitoring and evaluation of process variation. Using SPC Tools like various types of charts help in the trend analysis for the processes, analyze the vendors.

To report this post you need to login first.

4 Comments

You must be Logged on to comment or reply to a post.

  1. ram babu bharthepudi
    Aman,

    I think the information you have given in the blog is available in SAP documentation. I have couple of questions for you as I am trying to set up SPC. Is it possible for you to share your contact information?

    Cheers,
    Ram

    (0) 
    1. aman singh Post author
      Ram

      I am travelling these days, I can answer your queries sometime in 3rd week of Feb or you can drop me a mail if you want.

      Regards

      Aman

      (0) 

Leave a Reply