Visualizing process data

Learning outcomes

• Understand when it is appropriate to use scatter plots, bar plots, pie charts (hint: almost never), and even tables.
• Learn an interesting, potentially new plot: the box plot, to summarize and compare data.
• How to effectively visualize up to 5 dimensions on a 2-D plot, as shown in a video by Hans Rosling.
• Know the meaning of words like sparklines, data density, and chart junk.

Extended readings

• Sankey diagrams for example, would make a great way to show energy utilization in your company, or even a mass balance superimposed on a flowsheet. Here's a great example applied to the UK energy supply and demand.
• How To Lie With Financial Statistics, Investopedia, November 2011
• 40 years of boxplots
• Why you should never have to use pie charts, an article by Stephen Few.
• This is one video you must watch for the course: Hans Rosling shows an incredible data visualization

Resources

• Class notes 2015
• Class notes 2014
• Textbook, chapter 1
• Quiz and Solution
• Complete steps 1, 2 ... 9 of the software tutorial

Software codes for this section

Code to show how to superimpose plots

Try this code in a web-browser

# Run this code line-by-line (copy & paste) to understand the demonstration

data <- read.csv('http://openmv.net/file/raw-material-properties.csv')
summary(data)

# Single plot
plot(data$density1)

# Connect the dots
plot(data$density1, type='b')

# Another variable
plot(data$density2, type='b', col="red")

# Superimpose them?
plot(data$density1, type='b', col="blue")
lines(data$density2, type='b', col="red")  # where's density2 ?

# Superimpose them: limits
plot(data$density1, type='b', col="blue", ylim=c(10, 45))
lines(data$density2, type='b', col="red")  # now density2 shows up