Control Charts Using ggplot2 facet wrap R

问题

I have the following graph:

this was created using the dplyr group_by and summarise functions with ggplot2:

slopes %>%
   head(12) %>%
  inner_join(word_month_counts, by = "word") %>%
  mutate(word = reorder(word, -estimate)) %>%
  ggplot(aes(month, prop_per_month, color = word)) +
  geom_line(show.legend = FALSE,lwd=1.3) +
  geom_smooth(se=FALSE,lty=2)+
  facet_wrap(~ word, scales = "free_y")

I want to replace this with control charts and I have looked here and here but can't seem to workout how to incorporate when using facet_wrap

I have been playing with the qcc and qicharts like:

library(qicharts)

Datetime <- c("2015-09-29AM", "2015-09-29PM" ,"2015-09-30AM", "2015-09-30PM", "2015-10-01AM" ,"2015-10-01PM" 
              ,"2015-10-02AM", "2015-10-02PM" ,"2015-10-03AM" ,"2015-10-03PM", "2015-10-04AM" ,"2015-10-04PM" 
              ,"2015-10-05AM", "2015-10-05PM", "2015-10-06AM" ,"2015-10-06PM")
FailRate_M1 <- c(5045,4350,4350,3975,4290,4430,4485,4285,3980,3925,3645,3760,3300,3685,3463,5200)

df1 <- data.frame(Datetime,FailRate_M1)

qic(FailRate_M1,               
    x        = Datetime,              
    data     = df1,                  
    chart    = 'c',
    runvals  = TRUE,               
    cex      = 1.2,
    main     = 'Measurement Fail Rate (M1)', 
    ylab     = 'MFR (%)',          
    xlab     = 'Datetime')

Any pointers or code example with ggplot2 facet_wrap would be highly appreciated

回答1:

You can try:

# as you provided no reproducible example, I added afactor gr for the facet.
df1 <- data.frame(Datetime,FailRate_M1, gr=gl(2,8) )

# calculate the cl, ucl and lcl per group
df2 <- lapply(split(df1, df1$gr), function(data){
  a <- qic(FailRate_M1,               
           x        = Datetime,              
           data     = data,                  
           chart    = 'c',
           runvals  = TRUE)
  cbind.data.frame(ucl=a$ucl,cl= a$cl, lcl= a$lcl)
})

# the final data.frame  
df3 <- cbind(df1, do.call(rbind,df2))  

# and the final plot
ggplot(df3, aes(x=Datetime, y=FailRate_M1, group=gr)) + 
  geom_line(show.legend = FALSE,lwd=1.3) +
  geom_line(aes(x=Datetime, y=ucl)) +
  geom_line(aes(x=Datetime, y=cl)) +
  geom_line(aes(x=Datetime, y=lcl)) +
  facet_wrap(~ gr, scales = "free_x")

回答2:

You can use the 'tcc' function within the "qicharts" package.

Amongst the function arguments are g1 and g2 (grouping vectors used to define facets).

From the package function help:
"tcc() is a wrapper function for ggplot2() that makes multivariate run and control charts. It takes up to two grouping variables for multidimensional trellis plots".

For facet_grid() style plot you will need to specify both g1 and g2, else for facet_wrap() just specify g1.

Here are some slightly modified examples from the function help, and a couple of the resulting plots:

library(qicharts)

# Build data frame for examples
df <- data.frame(x = rep(1:24, 4),
                ReportMonth = (rep(seq(as.Date('2014-1-1'),
                              length.out = 24,
                              by = 'month'),
                          4)),
                num = rbinom(4 * 24, 100, 0.5),
                denom = round(runif(4 * 24, 90, 110)),
                grp1 = rep(c('g', 'h'), each = 48),
                grp2 = rep(c('A', 'B'), each = 24))

# Run chart with two grouping variables
tcc(num, denom, ReportMonth, g1 = grp1, g2 = grp2, data = df, main = "trellis run chart")

# C chart
tcc(num, denom, ReportMonth, g1 = grp1, g2 = grp2, data = df, chart = 'c', 
    main ="trellis C chart")

tcc(num, denom, ReportMonth, g1 = grp1, data = df, chart = 'p', 
       main ="trellis P chart , 1 facet")

# P chart
tcc(num, denom, ReportMonth, g1 = grp1, g2 = grp2, data = df, chart = 'p', 
    main ="trellis P chart")

# P chart with baseline fixed to the first 9 data points
tcc(num, denom, ReportMonth, g1 = grp1, g2 = grp2, data = df, chart = 'p', freeze = 9, 
    main = "trellis P chart with limits fixed to first 9  data points")

# U chart with two breaks and summary output
tcc(num, denom, ReportMonth, g1 = grp1, g2 = grp2, data = df, chart = 'u',
    breaks = c(9, 15), main =" trellis U chart with multiple breaks",
    print.summary = TRUE)

Note that the tcc function can produce run charts as well as control charts.
You can also modify the plot appearance with additional ggplot2 code.

Alternatively take a look at the ggQC package, which will also produce faceted plots easily ggQC.

Finally, a new version of qicharts2 qicharts2 has been released.

There is a facets argument within the qic function (the tcc function does not exist in the new package), so your code would have

facets = grp1 ~ grp2

within the function call.

来源：https://stackoverflow.com/questions/45548089/control-charts-using-ggplot2-facet-wrap-r

标签

plot

ggplot2

qcc