Streaming Analytics Value Chain: Multi-Stage Stream Processing Drives Value in IoT

To better understand the powers of Streaming Analytics, let’s look at one use case – preventing and mitigating power outages on the Smart Grid due to severe weather (e.g., the recent Northeast blizzard).

What if we could monitor the weather events and the electrical equipment, and predict in real-time when a piece of equipment was about to fail? We could then take defensive measures to prevent the failure or lessen its impact. In the extreme, if a catastrophic failure is predicted, then electric service could be automatically shut down to protect the equipment.

This is how we solve this problem using Streaming Analytics:

Streaming Analytics Value Chain

  1. The first step is to ingest the Smart Grid data at high speeds and volume. A large Smart Grid can generate up to 1 million events-per-second.
  2. The next step is to correlate all of the contextual data needed for advanced analytics, such as equipment profiles, recent maintenance history, grid topology, etc.
  3. And then correlate real-time weather conditions and near-term forecasts of severe weather to provide real-time situational awareness of what is happening to the grid right now.
  4. The next step is to predict “at-jeopardy” equipment and likely “imminent failures” using predictive failure models that are based on machine learning over historical data.
  5. The final step is to apply prescriptive analytics to determine the next best action to take – whether it is to send a work crew member or shut down the piece of equipment. And then, most importantly, to actually take the action in time to minimize the threat.

From data ingestion to automated corrective action, all of this needs to happen in less than 1 second!

We often speak of actionable intelligence, but it is intelligence with action that generates the value in IoT.

The number of stages in the pipeline can vary, depending on the use case, but it is generally true that it is not until the later stages, where you apply predictive analytics, prescriptive analytics, and actually take action that significant value is generated. Streaming Analytics facilitates this critical “last mile” in the IoT analytics value chain by enabling automated action to be taken.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>