Seeing, understanding, perceiving insights
Brief start about perception and understanding
Idea derives from the Greek word idein meaning “to see”. Question derives from Latin quaestiōnem meaning “to seek’”. Literally speaking, we seek to see; it is in our nature and without it, we can’t survive.
Also note that we say “ I see” when we understand something whereas " I hear, I smell" are also available proving that visual perception is the most predominant sense in us.
Besides, we might use data as mentioned in our article “Data types in Industrial visualization (code192.com)”, as the basis of reasoning to act and decide. Yet some forms of presentation such as visualized data emphasize our cognition and clearly communicate. If effectively designed, it provides clear insights from decisions that impact an area of promotions, investments, scheduling, planning, annual budgets … interventions needed to be made on production lines, and which team to choose for those interventions.
Rollback in History
As we mentioned earlier, visualizing data is nothing else but a form of communication. Let's fastly recap our history of communication with 2 major milestones.
The first one should be, as mentioned by Prakash Chakravarthi in “ The History of Communications from Cave Drawings to Mail Messages”,  drawings on rock and cave walls are the earliest methods of communication which we know [...] helping us to convey ideas and past events to other people.
In this article, we will discuss the necessity and reason for visualizing data, and give advice to improve the design.
Figure 1. Cueva de las Manos, Perito Moreno, Argentina. Hand marks dating between 7300 and AD
Yet, all systems have weaknesses as we can prove by trying to draw the message “why”. In fact, it is too abstract to be drawn. That is how we evolved our communication systems to prevail leading us to invent letters.
Writing is a way of visualizing data just like drawings as an improved communicating system helping us to illustrate our ideas. Letters are graphic symbols used to write. In fact, in “A History of Writing”, Seven Roger Fischer states that , In the ancient Middle East, around six thousand years ago, Sumer’s expanding society somehow had to administer and manage its incomes and expenditures [...] marking possession, an important part of book-keeping, probably provided some of the world’s earliest graphic symbols.
Figure 2. A large cuneiform inscription was found in Eastern Turkey around 2500 BC marking a message from the Persian king Xerxes. see ref
Just like drawings and writing, around the 1980's reports and dashboards started to combine these both in order to visualize data.
Figure 3. A dashboard example applying color based on labeled context.
Dashboards and reports communicate insights to us, just like cave drawings, cuneiforms, encyclopedias, books, or military reports,
Let us explore why we would actually need this system and how it could be a natural need to visualize industrial data.
This section covers the core need and inevitability of visualizing data.
According to Colin Ware in “ Information Visualization: Perception for Design “, we should be interested to visualize data  because the human visual system is a pattern seeker of enormous power and subtlety.
A list of some notions used in the upcoming sections.
Mental Discharge: the relief sensation we have compared to a previous experience.
Cognitive emphasis: an accelerated understanding of a subject.
Principle: is a comprehensive and fundamental law, doctrine, or assumption.
Gestalt: from German meaning “ form” is a school of psychology assuming that we see things as a whole rather than focusing on smaller components.
Data is abstract and finding insights in abstraction is not as evident for the whole of our audience.
Quick question Why would we need insight?
Quick answer To decide better and quicker in order to gain competitive advantage.
Figure 4. A dashboard example applying color based on labeled context.
In figure 4, it might be pretty time-consuming and difficult to acquire insights.
Thus, visualizing data as shown in figure 3 becomes much more presentable and less abstract.
We will discuss time-saving, and cognitive discharge, in the upcoming sections as the main reasons for visualizing data. In the next section, we'll discuss a way of improving these points.
Reading reports, and gathering information are crucial for decision making. Yet it is difficult to track many teams, production lines, or factory units as we quickly exceed the assumption of “ a person can retain an average 7 current elements in memory”. Let us compare a text-based insight discovery vs a visualized data example.
Imagine the following text-based report:
This report aims to illustrate our production compared to the last 15 days request. We produced a maximum of 14 thousand bottle caps which was on day 10. Our production did not meet request requirements on the last days 11,9,6,5 and 4. The largest gap between request and production occurred on day number 5. The minimum gap occurred on day number 11. Thus, we conclude to request a full analysis of events that occurred on day 5 to detect gap reasons.
Now let's compare this textual report to a visualized form
Figure 5. A widget example extracted from a dashboard
Figure 6 illustrates the same scenario as the textual one. Yet we can immediately compare the cognitive or mental discharge.
When reading figure 5, we get the message in a single glance: ask ourselves what happened on day 6 to intervene and prevent gap differences in upcoming scheduling if possible. We even get additional information: the fluctuation of requests and production between the last 15 days. Secondly, it is easier to compare thanks to the shape and color contrast supported with axes.
In the section, we will show off time savings.
Figure 6. A widget example extracted from a dashboard
When designed effectively, we save communication time when presenting our message to our audience. Figure 6 is an example of using gauges to evaluate if the actual water volume in our reservoir is in the given tolerance range for each section.
As you can see, our widget is supported with colors delimiting the “good and the bad” whereas the arrow demonstrates if we are “ good or bad”. Moreover, we avoid sending a team to each different section of our reservoir to observe these values. The data and insight are “compressed” as much as they should be, no more nor less in a single tiny widget.
Also, imagine we would need to check these values every day. It is less time-consuming to observe it in this kind of widget which even updates the values dynamically!
Design Application: Gestalt Principles
This section covers some principles applied to improve mental discharge, and effective communication when visualizing data through Gestalt principles.
According to Stephen Few in “Information Dashboard Design: The Effective Visual Communication of Data”, these principles  offer several useful insights that we can apply directly in our dashboard designs.
We visually group items, and objects next to each other as if they belong to the same cause.
Figure 7. A dashboard example with two separate “groups”
Figure 8. Same dashboard in figure X, except with all widgets within a single group.
If we would present these examples to an audience, visually, they would not perceive the same meaning.
Figure 9. Treemap widget example
We perceive objects as if they belong to the same group when they are bordered by distinct marks such as lines, rectangles, or color hues.
In this given example, the countries tend to group cities inside the parent rectangle.
Figure 10. Table grid widget example extracted from a report.
Even if our lines are striped in figure 10, we tend to perceive the content as a whole. Moreover, in this example, we apply alternating colors to ease reading the rows.
A step further: collaboration and interaction
Let's “add a cherry to our cake”, a french expression literally meaning: “ add something delicious to something already delicious”. Most visualization tools offer collaboration and dynamic interactions. So we can even gain more time and insight by taking advantage of these latest improvements.
We can conclude that visualizing data is inevitable if we work in factories using industrial data for the following reasons:
Fast and effective insight acquisition.
Visual perception is the dominant sense.