Hi all,
I am using SPC techniques of plotting Xbar, R, S, P and C charts in R.For
this I am using qcc package in R and also using the following document for
reference.
http://www.stat.unipg.it/~luca/Rnews_2004-1-pag11-17.pdf
But while plotting S chart, R chartchart and also C chart I am getting
following errors.
I have plotted the Xbar chart successfully using following command
obj <- qcc(a[,],type="Xbar")
But when I use the same command to plot the R chart and S chart jus by
replacing the type="Xbar" by type="R" and Type="R"
,I am getting folloiwing
error
Error in limits.R(center = NA_real_, std.dev = NA_real_, sizes = c(1000L, :
group size must be less than 51 when giving nsigmas
In addition: Warning message:
In sum(sizes * statistics) : Integer overflow - use sum(as.numeric(.))
obj<-qcc(D[trial],sizes=size[trial],type="p")
Similarly I am able to plot P chart for defective dataset successfully
using following command
y<-read.table("Defect_data_2008.csv",sep=",",header=TRUE)
attach(y)
obj<-qcc(D[trial],sizes=size[trial],type="c")
But while plotting C chart I am getitng following error
Error in plot.new() : figure margins too large
Could u please help me in solving these error.
Thanks in advance.
--
View this message in context:
http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1467901.html
Sent from the R help mailing list archive at Nabble.com.
Please add an reproducible example, so that we can see where it goes wrong. Maybe the error is raised by faulted data? Bart -- View this message in context: http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1468441.html Sent from the R help mailing list archive at Nabble.com.
ok. I will give the example for which i m getting this error. The data for
plotting R chart and S chart is very huge. SO i will take example for
plotting C chart.
Data is a follows :-
sample D size trial
1 11 1000 TRUE
2 11 1000 TRUE
3 13 1000 TRUE
4 11 1000 TRUE
5 14 1000 TRUE
6 12 1000 TRUE
7 17 1000 TRUE
8 7 1000 TRUE
9 7 1000 TRUE
10 6 1000 TRUE
11 9 1000 TRUE
12 3 1000 TRUE
13 2 1000 TRUE
14 4 1000 TRUE
15 13 1000 TRUE
16 11 1000 TRUE
17 7 1000 TRUE
18 11 1000 TRUE
19 8 1000 TRUE
20 5 1000 TRUE
21 37 1000 TRUE
22 11 1000 TRUE
23 9 1000 TRUE
24 4 1000 TRUE
25 7 1000 TRUE
26 4 1000 TRUE
27 17 1000 TRUE
28 11 1000 TRUE
29 18 1000 TRUE
30 21 1000 TRUE
31 18 1000 TRUE
32 19 1000 TRUE
33 12 1000 TRUE
34 11 1000 TRUE
35 13 1000 TRUE
Here my sample size is 1000 for each sample no.
1st column corresponds to sample no.
2nd column corresponds to no. of defects
3rd column corresponds to sample size
4th column , I have kept asTRUE because I want to consider whole data for
calculating the control limits.
The code which i use is as follows :-
library(qcc)
y<-read.table("Defect_data_2008.csv",sep=",",header=TRUE)
attach(y)
obj<-qcc(D[trial],sizes=size[trial],type="c")
But while plotting C chart I am getitng following error
Error in plot.new() : figure margins too large
For plotting the pchart if I just change the type="p" then it is
plotting
correctly without error.
--
View this message in context:
http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1468451.html
Sent from the R help mailing list archive at Nabble.com.
I suspect that you may have set qcc.options("cex") to
too large a value. Try lowering it with
qcc.options(cex=whatever)
-Peter Ehlers
vikrant wrote:> ok. I will give the example for which i m getting this error. The data for
> plotting R chart and S chart is very huge. SO i will take example for
> plotting C chart.
> Data is a follows :-
> sample D size trial
> 1 11 1000 TRUE
> 2 11 1000 TRUE
> 3 13 1000 TRUE
> 4 11 1000 TRUE
> 5 14 1000 TRUE
> 6 12 1000 TRUE
> 7 17 1000 TRUE
> 8 7 1000 TRUE
> 9 7 1000 TRUE
> 10 6 1000 TRUE
> 11 9 1000 TRUE
> 12 3 1000 TRUE
> 13 2 1000 TRUE
> 14 4 1000 TRUE
> 15 13 1000 TRUE
> 16 11 1000 TRUE
> 17 7 1000 TRUE
> 18 11 1000 TRUE
> 19 8 1000 TRUE
> 20 5 1000 TRUE
> 21 37 1000 TRUE
> 22 11 1000 TRUE
> 23 9 1000 TRUE
> 24 4 1000 TRUE
> 25 7 1000 TRUE
> 26 4 1000 TRUE
> 27 17 1000 TRUE
> 28 11 1000 TRUE
> 29 18 1000 TRUE
> 30 21 1000 TRUE
> 31 18 1000 TRUE
> 32 19 1000 TRUE
> 33 12 1000 TRUE
> 34 11 1000 TRUE
> 35 13 1000 TRUE
>
> Here my sample size is 1000 for each sample no.
> 1st column corresponds to sample no.
> 2nd column corresponds to no. of defects
> 3rd column corresponds to sample size
> 4th column , I have kept asTRUE because I want to consider whole data for
> calculating the control limits.
> The code which i use is as follows :-
>
> library(qcc)
>
y<-read.table("Defect_data_2008.csv",sep=",",header=TRUE)
> attach(y)
> obj<-qcc(D[trial],sizes=size[trial],type="c")
> But while plotting C chart I am getitng following error
> Error in plot.new() : figure margins too large
>
>
> For plotting the pchart if I just change the type="p" then it is
plotting
> correctly without error.
>
>
--
Peter Ehlers
University of Calgary
I'm sorry, I'm doing this in my spare time, hadn't time during the day. But I tried your example, and didn't get the error. If you get the error with your example (with the data you had sent to this list), then I think Peter is right, and it has something to do with margins you had set. Try to start a new R session, don't set your margins for new plots, and try if this works. Bart -- View this message in context: http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1469138.html Sent from the R help mailing list archive at Nabble.com.
I'm sorry, I'm doing this in my spare time, hadn't time during the day. But I tried your example, and didn't get the error. If you get the error with your example (with the data you had sent to this list), then I think Peter is right, and it has something to do with margins you had set. Try to start a new R session, don't set your margins for new plots, and try if this works. Bart -- View this message in context: http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1469147.html Sent from the R help mailing list archive at Nabble.com.
Thanks Bart and Peter for your help and the example is working for c chart as well withut any error. But when I am plotting the R chart i am still getting the following error. Error in limits.R(center = 62614.0571428571, std.dev = NA_real_, sizes c(1000L, : group size must be less than 51 when giving nsigmas I will just explain my example used for plotting R chart. I have around 35 groups and and each group has a sample size of 1000. I want to know whether sample size for each group chosen to be 1000 is a problem. Please help me. -- View this message in context: http://n4.nabble.com/problems-with-SPC-charts-in-R-tp1467901p1469800.html Sent from the R help mailing list archive at Nabble.com.