How to implement Control Charts in R?

Description

To know about the data types in R programming.

Need for Control Charts?

  • Control charts, also known as
    Shewhart charts or process-
    behavior charts, are a statistical process
    control tool used to determine if
    a manufacturing or business
    process is in a state of control.

 

Interpretation of Control Charts:

If it is in Control:

  • If all the points lie within the control limits,
    the process is said to be “in control”
  • If analysis of the control chart indicates
    that the process is currently
    under control, then no corrections
    or changes to process control parameters
    are needed.
  • Moreover, data from the method
    can be used to predict the future
    performance of the process.

 

If it is not in Control:

  • If the control chart indicates that
    the process is not in control,
    analysis of the chart can help determine
    the sources of variation, as this willresult in degradation of process
    performance.

 

Need of detecting Outliers:

  • Missing data are a common occurrence
    and can have a significanteffect
    on the conclusions that can
    be drawn from the data.

 

Package and Function:

  • R Package : qcc
  • R Function : groups(x) — For Grouping
    the data
  • R Function : qcc(x, type=, nsigmas=
    ,newdata= )
  • x –training
    data
  • type = “R”,”
    xbar” — Type of Control
    Chart
  • nsigmas — Default value is 3, means ±3
    standard deviation
  • Newdata — new sample data from the same
    process

#Control Charts
#Loading required packages
#install.packages(“qcc”)
library(“qcc”)
data(pistonrings)
attach(pistonrings)
head(pistonrings)

#Grouping the data
dia<-qcc.groups(diameter,sample)
head(dia)

#R Chart
obj<-qcc(dia[1:30,],type = “R”)
summary(obj)

#X- Bar Chart
obj<-qcc(dia[1:30,], type=”xbar”)
summary(obj)

#Changing the Values of nsigma
obj<-qcc(dia[1:30,],type=”R”,nsigmas = 1)
summary(obj)

#Testing new Data
obj<-qcc(dia[1:30,],type = “xbar”,newdata = dia[31:40,])

Leave Comment

Your email address will not be published. Required fields are marked *

clear formSubmit