) against these gate limitEach time you complete a data subgroup, the program can perform one or more of the following checks on that subgroup. If you are entering data with a subgroup size of one, the following checks are performed after each data value (each subgroup of one) is entered.
GainSeeker only performs the checks that you specify in the standard for this subgroup.
Limit values entered in the standard
Gate limits
If you have set gate limits in the standard, the program can check the subgroup average ( ) against these gate limit
Range gates
If you have set range gate limits in the standard, the program can check the subgroup range ( R ) against these range gate limits.
Failures based on the most recent data for this standard
When you open a data entry session and select a standard, the program retrieves the most recent data for that standard and uses that data set to generate the chart(s) on your screen and to calculate statistics on that data set, including the mean ( ), the average range ( 
), and upper and lower control limits. It uses this data set and the statistics based on that data to perform the checks listed below.
Outside control limits.
When you complete a data subgroup, the program can check the subgroup average ( ) and range ( R ) against the control limits calculated from the most recent data retrieved for this standard.
Run on control chart.
When you complete a data subgroup, the program can analyze the average ( ) for the subgroup you just completed and for several previous subgroups to make sure that there is not a significant number of subgroup averages in a row on one side of the mean ( 
).
The number of points that constitutes a run may be 6, 7 or 8, depending how you configure the program.
Run on range chart.
When you complete a data subgroup, the program can analyze the range ( R ) for the subgroup you just completed and for several previous subgroups to ensure there is not a significant number of ranges in a row on one side of the data average range ( ).
The number of points that constitutes a run may be 6, 7 or 8, depending how you configure the program.
On a range chart, GainSeeker anticipates that more values will plot below the R-bar than above it. These counts are automatically adjusted based on subgroup size. The following lengths are used:
Subgroup size of 1 or 2:
Above the mean:
The number of points in a run that plot above the R-bar = run length (as set in the System Administration module) - 1 (=7 if configuration is set to 8)
Below the mean:
The number of points in a run that plot below R-bar = run length (as set in the System Administration module) + 2 (= 10 if configuration is set to 8)
Subgroup size larger than 2:
Above the mean:
The number of points in a run above R-bar = run length (as set in the System Administration module) (= 8 if configuration is set to 8)
Below the mean:
The number of points in a run below R-bar = run length (from Administration + 1 (= 9 if configuration is set to 8)
Trend on control chart.
When you complete a data subgroup, the program can analyze the average ( ) for the subgroup you just completed and for several previous subgroups to make sure that there is not a significant number of subgroup averages in a row that are consistently increasing or consistently decreasing.
The number of points that constitutes a trend may be 6, 7 or 8, depending how you configure the program.
Trend on range chart.
When you complete a data subgroup, the program can analyze the range ( R ) for the subgroup you just completed and for several previous subgroups to make sure that there is not a significant number of ranges in a row that are consistently increasing or consistently decreasing.
The number of points that constitutes a trend may be 6, 7 or 8, depending how you configure the program.
2 of 3 subgroups outside 2 standard deviations.
When you complete a data subgroup, the program can analyze the average ( ) for the subgroup you just completed and for the two previous subgroups.
It will make sure that two of the three subgroup averages ( ) are not greater than 2 standard deviations above the mean ( ).
It will also make sure that two of the three subgroup averages ( ) are not less than 2 standard deviations below the mean ( ).
4 of 5 subgroups outside 1 standard deviation.
When you complete a data subgroup, the program can analyze the average ( ) for the subgroup you just completed and for the four previous subgroups.
It will make sure that four of the five subgroup averages ( ) are not greater than 1 standard deviations above the mean ( ).
It will also make sure that four of the five subgroup averages ( ) are not less than 1 standard deviations below the mean ( ).
CuSum failure.
When you complete a data subgroup, the program can check the upper and lower CuSum values ( QU(n) and QL(n) ) against the Acceptance limit calculated for this standard.
If the subgroup fails any of these real-time checks, the program will display a failure note screen that indicates which real-time check failed. For information on using the failure note screen, see Responding to real-time failures for the subgroup.