IoT applications for businessKevin Ashton, co-founder and executive director of the Auto-ID Center at MIT, explains how IoT can have an impact:“Today computers -- and, therefore, the internet -- are almost wholly dependent on human beings for information. Nearly all of the roughly 50 petabytes (...) of data available on the internet were first captured and created by human beings by typing, pressing a record button, taking a digital picture or scanning a bar code.The problem is, people have limited time, attention and accuracy -- all of which means they are not very good at capturing data about things in the real world. If we had computers that knew everything there was to know about things -- using data they gathered without any help from us -- we would be able to track and count everything and greatly reduce waste, loss and cost. We would know when things needed replacing, repairing or recalling and whether they were fresh or past their best.”Businesses can use IoT applications especially to collect data. If properly set up, data collected via IoT can be a lot more reliable as there’s no need for human intervention. Also, since it’s automated, the data collection process happens a lot faster.Data collected via IoT can be used to streamline and improve operations. Additionally, it can also improve overall business efficiency. Some of the most popular uses of data include predictive maintenance, inventory management and optimisation, and customer insights.Probably one of the industries that would benefit the most from IoT applications would be retail, logistics (warehouse management), and manufacturing. A Forrester study shows potential applications for various industries - see the image below. It also estimates that these will drive the larger part of IoT growth in the years to come.(Image source)
Building an IoT prototype for businessesGiven the potential applications for business, we decided to experiment with IoT inside Qubiz as well. So we built a prototype that focuses on collecting data such as temperature, humidity, and acceleration from multiple IoT devices equipped with sensors. Thus, our system captures, sends and analyses data in real time.Data collection in IoT is done via sensors. Any item that has the sensors installed can act as a data source/IoT device. To build our prototype, we used:
- Windows 10 PC/ Android device with Bluetooth
- Azure Cloud services (Event hub/IoT Hub, Stream analytics, storage accounts and Power BI)
- SimpleLink Bluetooth low energy/ Multi-Standard SensorTag from Texas Instruments.
How the IoT prototype worksOur prototype features a rather straightforward architecture. The sensors are installed on various devices in order to collect data such as temperature or humidity level.Next, the sensors connect via Bluetooth to the Android/Windows 10 device to send the data. Because the connection is made via Bluetooth, there are no bottlenecks in the sensor-Android/Windows 10 device communication. The Windows 10/Android devices act as a gateway for the IoT devices and send the data further to the Cloud (using a low latency protocol AMPQ) and into a data stream. From this stage, we can:
- archive the original data
- process incoming data via real-time analytics (through Power BI)
- send it to other data processing systems.
More details about our IoT proof of conceptOur IoT system was designed to support large quantities of data. It also performs heavy-duty processing operations fairly easily in real time in a flexible, scalable manner. These are performed via stream analytics services. Additionally, we’ve implemented multiple live stream services: Amazon, Microsoft, and Google. This took a bit longer as each implementation depends on provider specs, but it’s safer.To make it easily accessible, we used Windows as it’s one of the most popular operating systems. Much to our surprise, integrating Windows was pretty challenging as we discovered that it’s not the best option for some tasks such as data collection. It was also hard to get the Windows-powered device to act as a gateway system, to connect it to the IoT device and keep the connections open.Another significant challenge was being able to read data coming from a sensor. For instance, if the battery level drops below a certain level (currently 70%), the sensor tags stop sending messages and information. Also, the sensor tags battery level is not very accurate (it’s based on voltage and not amperage). We also had to experiment to find the best sampling intervals. If sampling intervals are too high, the IoT device has issues sending data.
This screenshot shows all the available sensors and the data they collect.
This screenshot features data from the LightIntensity sensor.
This screenshot features data from the Accelerometer sensor.
Data can also be viewed on PowerBi via desktop.
Here we can see data in real-time via PowerBI online analytics stream.