Scribd
Upload
7 views24 pages

L-2 Data Visualization

The document discusses the importance of process data analytics in chemical engineering, highlighting its compatibility with first principles models, ability to handle uncertainties, and generation of interpretable solutions. It covers various types of data, the significance of effective data visualization, and provides examples of good and bad plots. Additionally, it emphasizes the need for clear visual representations to aid decision-making and understanding of complex data trends.

Uploaded by

mohak.2025.2026
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views24 pages

L-2 Data Visualization

The document discusses the importance of process data analytics in chemical engineering, highlighting its compatibility with first principles models, ability to handle uncertainties, and generation of interpretable solutions. It covers various types of data, the significance of effective data visualization, and provides examples of good and bad plots. Additionally, it emphasizes the need for clear visual representations to aid decision-making and understanding of complex data trends.

Uploaded by

mohak.2025.2026
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 24

ChE 208: Process Data Analytics

& Monitoring
Data visualization -I

Department of Chemical Engineering, IIT Indore


What set process data analytics apart from other big data
analytics, such as business intelligence ?

1. Working compatibly with first principles models or process knowledge

2. Dealing effectively with uncertainties that are usually time varying

3. Generating interpretable solutions

Qin, S. J., & Chiang, L. H. (2019). Advances and opportunities in machine learning for process data analytics. Computers & Chemical Engineering, 126, 465–473.
https://doi.org/10.1016/j.compchemeng.2019.04.003

Department of Chemical Engineering, IIT Indore


What is Data ?
Data is any raw information, facts, or statistics that you can analyze and interpret.

Department of Chemical Engineering, IIT Indore


Types of data
Structured Data Unstructured Data

• Record data (measurements taken • Images


from a process) • Videos
• Data Matrix (numeric attributes) • Audio speech
• Graph Data (relational objects) • Long Text
• Ordered Data (Temporal / sequence • Nominal data
/ spatial )
• Discrete / continuous data

Department of Chemical Engineering, IIT Indore


Some examples …
➢Co-worker: Here are the yields from a batch system for the last 3 years
(1256 data points), can you help me:
• understand more about the time-trends in the past 3 year?
• efficiently summarize the yield from all batches run in 2020?

➢Manager: effectively summarize the (a) number and (b) types of defects on
17 aluminum grades for the past 12 months

➢Yourself: 24 different variables being measured vs time (5 readings per


minute, over 300 minutes) for each batch we produce; how can we
visualize these 36,000 data points ?

Department of Chemical Engineering, IIT Indore


Batch systems generate large quantities of
extremely valuable data

Department of Chemical Engineering, IIT Indore


Batch systems generate large quantities of
extremely valuable data

Department of Chemical Engineering, IIT Indore


We need good plots to make decisions
quickly, correctly, and confidently

Department of Chemical Engineering, IIT Indore


Why bother learning about this topic: it's too easy!
• This class might seem too easy; too obvious. It is!

• The human eye and brain are excellent at pattern recognition, sorting through signal and
noise.
• We can easily cope with bad plots; but good plots save time and show a clearer, more
honest picture.

• Cliches: \Let the data speak for themselves", \Plot the data”

• Strong suggestion: Find a bad plot (journal publications, an old lab report that you have
written); upload it to the GC assignment and criticize the plot. Why is it bad?

Department of Chemical Engineering, IIT Indore


Time-series Plots
• Show a univariate piece of information in 2 dimensions
• It can be used generally to display any sort of sequence, whether it is along time or some other
ordering

Department of Chemical Engineering, IIT Indore


An example of a bad time-series plot; what
problems can you identify?

Department of Chemical Engineering, IIT Indore


Notice how the plot's “message" is entirely
different now

A first attempt at fixing the prior visualization


Department of Chemical Engineering, IIT Indore
Poor plots; default settings in plotting
software create cluttered plots

Department of Chemical Engineering, IIT Indore


Use separate, parallel axes rather to compare
plots (and minimal data ink)

These non-default settings can take a long time to set (10 minutes for this example)

Department of Chemical Engineering, IIT Indore


Sparklines are a type of time-series plot
• Sparklines are small graphics that carry a high density of information
except, we omit the horizontal and vertical axes (they are implicit)

• Useful for financial trends


• Great for iPods, cell phones, tablet computers because they are of high
density and small size
• Our eye can detect 250 dots (points) per linear inch and 650 points per
square inch.
Department of Chemical Engineering, IIT Indore
Sparklines are used on various websites now
to show high-density graphics

Screenshot from Google Finance 08 January 2014


Notice how you clearly detect correlations
in stock prices (stocks that move together).

Department of Chemical Engineering, IIT Indore


Example of sparklines in everyday use

Department of Chemical Engineering, IIT Indore


Keep the x-axis spacing constant on time-series
plots: helps interpretation

➢ Use another plot (e.g.


below the original) to
zoom in on details

Department of Chemical Engineering, IIT Indore


Show a reasonable amount of historical data
for context

Department of Chemical Engineering, IIT Indore


Important learning points from time-series plots

➢avoid using colour as your message


➢ use honest scaling on your x-axis and y-axis
➢ the human eye (and brain) can deal with vast quantities of data:
exploit it

Department of Chemical Engineering, IIT Indore


Quiz time !
➢The data shown here are the number of visits to a university website
for a particular statistics course. There are 90 students in the class.

1. What are the names (type) of the 2 plots shown ?

2. List any 2 interesting features in these data.

Department of Chemical Engineering, IIT Indore


Department of Chemical Engineering, IIT Indore
Department of Chemical Engineering, IIT Indore
➢ Departmental expenses are
encoded as simple line graphs,
which beautifully present the overall
trend and individual ups and downs
of the values through time.

➢ This arrangement of eight graphs


within eye span, one per
department, sorted from the
greatest to least expenses, tells the
data’s story clearly.

Department of Chemical Engineering, IIT Indore

You might also like