procedure QCCumSumChart(const Data: TMtx; const DrawVec: TVec; k: Double; h: Double; mean: Double; sigma: Double; const SHigh: TVec; const SLow: TVec; Confidence: Double);
CUMSum chart.
| # | Name | Description |
|---|---|---|
| 1 | Data | Each Data row contains replicated observation taken at specific time. |
| 2 | Confidence | Confidence level for upper and lower control limit. Confidence must lie in the (0,1) interval. |
| 3 | DrawVec | Returns values to be drawn. |
| 4 | mean | Defines the estimate of the in-control mean. |
| 5 | sigma | Defines known (or estimated) standard deviation of the sample means. |
| 6 | k | Design parameter of the V-mask. k is the rise in the V-arm corresponding to one sampling unit. |
| 7 | h | Design parameter of the V-mask. h defines the rise in the arm coresponding to the distance from the origin to point vertex. |
| 8 | SHigh | If set, returns high values for cumsum plot. When SHigh exceeds value h, the process is said to be out-of-control. Compare SHigh values to h to find if a process is out-of-control. |
| 9 | SLow | If set, returns low values for cumsum plot. When SLow exceeds value h, the process is said to be out-of-control. Compare SLow values to h to find if a process is out-of-control. |
Result: stored in self (calling object)
Remarks:
Calculates CumSum Chart drawing values and additonal values, needed for deciding if some samples are out-of-control.
See www.itl.nist.gov/div898/handbook/pmc/section3/pmc323.htm to learn more about Cumsum QC charts.
Examples
var h,k,m,s: double;
data: Matrix;
sh,sl, s: Vector;
outofcontrol: VectorInt;
begin
data.LoadFromFile('process_data.mtx');
// estimate k=0.32, h = 4.7
k := 0.32;
h := 4.7;
// estimate process mean and sigma
m := 230.3;
s := 5.2;
QCCumSumChart(data,s,k,h,m,s,sh,sl);
// find point indexes which exceed h (out-of-control points)
outofcontrol.FindIndexes(sh,'>',h);
end;