Predictive analytics - understanding the future failure modes of the device

Predictive analytics has its roots in the ability to predict what might happen based on analysis of the sensor data, typically using machine learning algorithms, given a training dataset. These analytics are about understanding the future based on past data models. Predictive analytics typically provides the user with actionable insights based on past data. In IoT applications, these types of analytics can generate actionable alerts or advisories or KPI.

Predictive analytics, using machine learning, is, at its root, the basic practice of using algorithms to parse data, learn from it, and then make a determination or prediction about something in the future. So, rather than hand-coding software routines with a specific set of instructions to accomplish a particular task, the machine is "trained" using large amounts of data and algorithms which give it the ability to learn how to perform the task and provide estimates about the likelihood of a future outcome. Spark ML provides a nice infrastructure to develop/deploy predictive analytics. One of the successful analytics that has been in the market for IIoT is the SmartSignal analytics which is offered by GE Digital for many different industrial use cases.