Data collection is a key function of IoT projects. Data is used for analyzing trends, observing status, finding anomalies, and generating reports. Status data provides the current snapshot of a process or activity. Analyzing trends can reveal new opportunities to improve efficiencies or understand behavior. Finding anomalies can predict and prevent failures.
Dashboards are the most common way for operators, managers, and data analysts to visualize status and trends. There are hundreds of ways to represent data visually on a dashboard using charts and widgets. The type of chart or widget you choose depends on the target application of the data such as viewing status, studying trends, or finding anomalies. There are a few charts and widgets that are commonly used because of their simplicity and ease of understanding. In this post we review the top 5 charts and widgets that are most used in creating IoT dashboards.
A tile chart is useful for viewing the latest value of a data stream (e.g., “Tank Level”). The picture above shows a tile chart that uses text, images, and color to represent data. Images and color are a terrific way to draw attention to the data, especially when the dashboard is displayed on a large screen display at a distance.
Line and Area Charts
Line and area charts are useful for studying trends in data. Stacked line and area charts allow users to visualize multiple data streams on the same chart enabling users to study correlation between data streams or cause-effect relationships. Studying historical data is another application of line and area charts. For example, you can use a line or area chart to study temperature changes or battery level over time. Using a stacked line chart, you can study the correlation between temperature and battery level.
Tables are useful for summarizing and organizing data. Typically, each row displays the current value of a data stream. Tables are also used to display reports. In a report table, the columns display data values. The rows can be timestamps or data source names. A log is an example of a report where each row is associated with a timestamp. A registry table is an example of a report where each row is associated with a data source.
Heat maps are useful for viewing three-dimensional data in two dimensions (such as your monitor or printed paper). Color is used to represent the Z-axis data, where the Z-axis is perpendicular to the surface. Weather maps are a notable example of heat maps. Plotting the surface imperfections of a metal sheet is another example of heat maps.
Geographical map charts are useful for viewing data associated with geographic location. For example, an air quality sensor with an attached GPS has air quality data that is associated with geographic locations.
In the example above from www.purpleair.com, a real-time display of air-quality sensors is shown for a selected geographical area. Additional information such as sensor counts, values, and status can be overlayed on the map. Maps allow zooming in on the data and drilling into the information from a specific sensor.
There are many other types of charts (pie, bar, scatter, regression, sunburst, gauge…) that can be used to display the right data, in the right format, to the right users. Choosing the correct chart maximizes inference while minimizing confusion. Data dashboards continue to be one of the most used features of IoT software.