Analytics at the edge: examples of opportunity in the Internet of Things

The Internet of Things in energy

A detailed view of energy consumption patterns is needed to understand energy usage, daily spikes and workload dependencies. And beyond just manufacturing – lighting alone takes 19% of the world's electricity, said the International Energy Agency. Optimized alternate energy sources can have a significant impact on all sectors.

For example, a single blade on a gas turbine can generate 500GB of data per day, according to GE. Wind turbines constantly identify the best angles to catch the wind, and turbine-to-turbine communications allow turbine farms to align and operate as a single, maximized unit.

Historically, the only way to know what was happening with a turbine – even if it was on and working – was to climb 330 feet and see. Remote monitoring provides new eyes on the status of these energy generators.

What if the same data could be used to forecast? Efficiency in green energy means we can store more energy for use when the wind is low. Predicting when excess energy is available can help determine when to charge batteries, for example, further extending the efficiencies of alternate energy sources.

Of course, the energy market provides one of the most well-known examples of IoT technology altering the customer landscape. With dynamic smart meter billing, customers have new choices, which lead energy companies to adopt a more customer-centric approach. The utility smart grid transformation is expected to almost double the customer information system market, from $2.5 billion in 2013 to $5.5 billion in 2020, according to a study from Navigant Research.

The Internet of Things in retail

Customers are also at the center of IoT analytics in retail, where some companies are studying ways to gather and process data from thousands of shoppers as they journey through stores. This "in-store geography" informed by sensor readings and videos considers how long shoppers linger at individual displays, recording what they ultimately buy.

With the goal of optimizing store layout, these data points can also be tied to smart-device Wi-Fi networks. In addition to appropriately targeting shoppers for promotions in-store, retailers can ask customer opinions – using IoT data to initiate an interaction, customizing the shopping experience and enhancing loyalty.

Taking action with IoT data

Event streams monitor patterns of interest. Sensors and devices generate lots of data that describes existing conditions. Analysis of conditions informs what actions are necessary – either immediately as an alert notification, or with pre-planning from predictive and other advanced analysis methods.

Of course, analysis always leads to more questions – directing what additional sensors (and data) can be collected to measure new aspects of the conditions, elements of the event or more detail in the scenario to understand different patterns.

IoT data, by itself, isn't the value. Just as with traditional data sources, it's the ability to take the insights and then act on them that provides value. To know what to do in the moment, use analytics at the edge.

  • Fiona McNeill is Global Product Marketing Manager at SAS