It is a strange and inaccurate question to ask; what do you want to eat yesterday? However, IoT systems are architected to often ask very similar questions by aggregating a massive amount of data in a central repository where every so often a data scientist can evaluate and perhaps realise some insights that would have been best realised and responded to at the time the events were taking place at the edge.
Guest blog by Kevin Konkos.
To be clear, there are a number of valuable insights that can result from big data acquisition and analysis, and the outcomes of those insights should be considered in developing an IoT system. But there is yet another class of questions and insights that are best asked in real-time.
With this in mind, Wind River is expanding its VxWorks RTOS IoT product portfolio by introducing the new analytics product, AXON Predict Analytics for VxWorks, providing our VxWorks developers a way to embed on-device analytics into their applications to take advantage of predictive analytics, machine learning, and artificial intelligence.
AXON Predict Analytics processes data without the need for cloud connectivity so our VxWorks developers using AXON Predict Analytics can now manage critical data at the edge of a network through visual edge analytics software and get autonomous responses in real time.
Figure 1: AXON Predict Analytics for VxWorks architecture diagram
Best of both worlds – stop the hacking while decreasing costs
We all know that one reason IoT growth is being propelled is by the ever decreasing costs of network infrastructure, but for customers those costs still do remain a key overall system cost driver that can be optimised with analytics at the edge. In addition, by utilising increased on-device compute capabilities with our new analytics, our customers will get the added benefit of being able to limit unnecessary data propagation - decreasing the risk of their systems from being hacked.
Which we all know is so important these days. The new VxWorks analytics product provides these sharper insights and faster response times while managing both costs and security risks to help customers improve their top and bottom lines. It’s the best of both worlds.
Predictive maintenance – analytics proves its value
While the use cases for AXON Predict Analytics are almost infinite, there is immediate impact and high business value for customers in particular for predictive maintenance, self-healing, safety and security applications. According to an article on TNW, regular preventative maintenance on a ten year-old air compressor worth $32,900 can extend the machine’s life for up to four years, resulting in a total saving of up to $6,359.
Think about other expensive difficult to maintain devices and especially ones in locations with limited network connectivity such as drills on oil rigs, airplanes, military vehicles, wind turbines located in deserts, cranes, mining equipment, etc. For these types of complex disconnected devices, edge analytics can easily prove its value.
Courtesy of Wind River.