Here we will explore a Qlik Sense app that analyzes IoT data. There is “much ado” made of the new technology trend, the Internet of Things, otherwise known as IoT. Google defines the Internet of Things as
The interconnection via the Internet of computing devices embedded in everyday objects, enabling them to send and receive data.
Statista predicts that by 2020, IoT embedded devices will surpass 50 billion units. So this is obviously a trend that is here to stay.
They Don’t Need Us
We are used to our personal devices that provide directions or relay the news to us. Most of the devices in our life are in our possession or otherwise assist us directly.
The IoT does not want or need us.
In IoT networks, devices speak to each other or to centralized servers. Their sole mission is to collect and transmit very limited and specific data. This gets stored into large databases and then automated algorithms take over to adjust settings or the data is mined and analyzed by AIs or possibly humans. IoT sensors are currently being put in all sorts of places including:
- Your home thermostat
- Parking meters
- Water and other resource management facilities
- City streets and highways
- Embedded in your neck (just kidding about that one…or am I?)
OK, Then What
This mass data collection will create great need for quick and useful tools to process and analyze the data, which is where Qlik Sense fits in. Platforms that can provide focused solutions that solve problems of volume, automated analysis and timeliness (ie. real-time analysis) will be the leaders in this space. Where does Qlik Sense fit into the IoT?
Qlik Sense and IoT
I thought it would be interesting to take some IoT data and create a Qlik Sense app from it. This does not do much to solve the challenges I mentioned, but hopefully this will at least give us a glimmer of what is possible with this kind of data.
City Pulse – Traffic Monitoring and Infrastructure Analysis
The partners of the CityPulse EU FP7 project collected and published several IoT datasets for the public for analysis projects. I pulled this data down to build out this application. Hundreds of road sensors were installed in Aarhus, Denmark. They monitor passing cars and speed between observation points. Although there were several months of data available, I pulled only one set for this application. The central table contains only 4 million records. Traffic information is incredibly tangible to everybody who has been in a car and allows me to feature “pretty maps” so I found it to be a great use-case for IoT analysis. Here we will just do screenshots and commentary of some of the sheets. Feel free to download the application to further explore the use-case. Note, most of the maps will not be available unless you have Qlik GeoAnalytics. IoT City Pulse.qvf I was thinking one of the use-cases for this data would be to alert people when a sensor was down or there was a stop in traffic. The points that had no records for the latest time have been isolated. The second part of the alert screen displays the total traffic volume for the day and then trended by hour. The next sheet analyzes the overall traffic. The heat map below shows the areas of the city with the most traffic. I imagined that a user might want to look at one particular stretch of road. This sheet allows users to select a stretch and then look at its traffic flow and statistics. There is a great option in Qlik GeoAnalytics that lets you add an arrow when a start and end point are utilized. This sheet allows users to compare days of the week, slice data by AM/PM and examine hourly traffic. The colors of the radar chart were poorly chosen here (by me), but it sure does look cool. The last sheet examines the disparity between directions of traffic. Why does the traffic flow smoothly in one direction but not the other? Identifying roads with the highest disparity is the starting point. Then we can examine individual points on the map to possibly come up with reasons why.
The app is available for you to download here: IoT City Pulse.qvf Note, most of the maps will not be available unless you have Qlik GeoAnalytics. The data sources can be found here: CityPulse Dataset Collection In addition to Qlik Sense, I also utilized GeoAnalytics (requires a license) and several free extensions: Line Widgets – Thank You Hector Munoz! Interactive Radar Chart Simple KPI
As I said earlier, there are many real challenges in dealing with Internet of Things data as it relates to Qlik Sense. I do believe we have all the tools but it would be worthy of a deeper explanation. With that said, my mind was reeling from the possible applications of this kind of data in all sorts of use-cases. As these sensors continue to decline in price, I can only imagine the ramifications in our world. What are your thoughts on the Internet of Things? Is it a fancy buzzword or is there something deeper? Happy Qliking!