Why Edge Analytics should be part of the IOT strategy of an Organization

Sridhar krishnan

Sridhar krishnan

The organizations in today’s ever changing competitive world should be capable of adopting to changes quickly and seamlessly. The organizations should carefully invest in right technologies to align with its strategy to attain maximum benefit. With growing and increasingly disbursed sources of information and the pace of organizational change accelerating rapidly, the ability to filter and analyze only the time-sensitive data both in real-time and historical in edge side and non-time sensitive data in the cloud is invaluable. It is cost effective approach to have central data analytics infrastructure only for non-time sensitive analysis and moving analytics to Edge gives an opportunity to monitor and running stream and batch analytics to get insights to take decisions quickly with speed and simplicity.

Three central value propositions of Edge Analytics:

Real-time response – There are many critical systems that can’t depend on a cloud connection for a decision. A few seconds of latency can make a significant difference for an operation with massive power consumption from multiple sources. And sometimes connections fail entirely, which is far more damaging than simple latency.

Cost of data transmission – Rule-based engines can filter out the noise and send only the interesting information back to the central repositories. Gateways can also batch information into packets with smaller footprints that are more optimal for a given means of transmission.

Information management – Edge processing can make a significant impact by cleaning and be harmonizing data before sending it off to central information management systems. this type of edge processing improves central data mining and analytics capabilities.

Edge Analytics is the game changer in all the industries including industrial IOT, retail, manufacturing, finance, energy and agriculture sectors.

Core benefits of edge analytics and industries where it makes more sense are,

  • Remote Monitoring in Oil & Gas Drilling, Oil & Gas Refineries, and Wind Turbines industries.
  • Preventive Maintenance in Factory Robots, Airplane Tires, and Energy Grid industries.
  • Personnel Safety in Refinery Gas Leaks and Contamination Containment.
  • Real-time Quality Assessment in Oil Drilling, Manufacturing Cell, Train Repair
  • Asset Health in Readiness Assessment.
  • Efficiency Through Digitization & Automation in Smart Meters, Utility Billing, etc.
  • Cost Reduction Through Better Facilities Management in Energy Management & Reduction.

Business use case examples

Manufacturing organizations must run data aggregation, data preparation and analytic workflows at the source, where the data is generated. When linked to sensors along the path of production, analytic models running on gateway devices at the network edge evaluate and score manufacturing output by parameters such as size, temperature, pressure, color, vibration, and weight. When variances are detected, the embedded sensor, smart device or gateway immediately sends alerts or even stops production to help limit waste. The manufacturing process benefits from the flexibility to monitor and update the model at the point where data is generated and from the network efficiency to send only valuable information like state changes back to the cloud for deeper analysis.

By using machine learning, data mining and advanced analytics at the source of the data to examine thousands of steps per process, manufacturers can catch small, bad batches before they become large, seriously bad batches. One pharmaceutical company discovered that minimizing scrap and wasted resources saved them several hundred thousand dollars. Their Edge analytics implementation combined with a program of statistical process control featuring audit trails and role-based security, allowed them to recover their investment in edge analytics in the first quarter after implementation.

semiconductor manufacturers automatically analyze and classify patterns of failures such as scratches or defects around the edges of silicon wafers. They identify possible root causes, and the specific processing steps and respective tools and machines that require inspection or maintenance.

Retail customer behavior analysis
Near instant edge analytics on sales data, images, coupons used, traffics patterns, and videos are created – provides unprecedented insights into customer behavior. This intelligence can help retailers better target merchandise, sales, and promotions and help redesign store layouts and product placement to improve the customer experience. One way this is accomplished is through the use of edge devices such as beacons, which can collect information such as transaction history from a customer’s smartphone, then target promotions and sales items as customers walk through the store.

Edge Analytics helps banks to understand their customers better by providing insights such as location-based suggestions and customer recommendations. Embedded in the bank’s customer channels – online banking or mobile banking, edge analytics delivers transactional behavior and location-based suggestions in real time.

We believe Edge Analytics can come into play in sectors such as farming and agriculture largely wherein regardless of the network, analytics can point out equipment failure or irrigation leaks.


Machine Learning and IoT


Srinidhi Murthy

Given all the hype and buzz around machine learning and IoT, it can be difficult to cut through the noise and understand where the actual value lies. In this Blog, we explain how machine learning can be valuable for IoT when it’s appropriate to use, and some machine learning applications and use cases currently out in the world today.

What is Machine Learning?

Machine Learning is not a novelty innovation. As early as 1959, Arthur Samuel defined the concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do. Of course, the timeline from definition to implementation in everyday life can be a long one. Today, many factors have come together to make machine learning a reality, including large data sources that are great for learning, increased computational power for processing information in split seconds, and algorithms that have become more and more reliable.

What is Data Analytics? How is it different from Machine learning?

Data analytics can help quantify and track goals, enable smarter decision making, and then provide the means for measuring success over time.

Machine learning, on the other hand, is a process of continuous learning, to which the system can make immediate adjustments to improve processes, timelines, decision making etc.

Machine Learning Use Cases in IoT

The data models that are typical of traditional data analytics are often static and of limited use in addressing fast-changing and unstructured data.

When it comes to IoT, it’s often necessary to identify correlations between dozens of sensor inputs and external factors that are rapidly producing millions of data points.

In general, machine learning is valuable when you know what you want but you don’t know the important input variables to make that decision.

Some of the typical use cases of Machine Learning and IoT are given below.

Cost Savings in Industrial Applications

Predictive capabilities are extremely useful in an industrial setting. By drawing data from multiple sensors in or on machines, machine learning algorithms can “learn” what’s typical for the machine and then detect when something abnormal begins to occur.

A Large Equipment Manufacturer has installed Many IoT sensors on its equipment which continuously send data to be learned and any deviation above a threshold be highlighted and immediately triggered as a notification to the concerned person.

Shaping Experiences to Individuals

We’re actually all familiar with machine learning applications in our everyday lives. Both Amazon and Netflix use machine learning to learn our preferences and provide a better experience for the user. That could mean suggesting products that you might like or providing relevant recommendations for movies and TV shows.

Similarly, in IoT machine learning can be extremely valuable in shaping our environment to our personal preferences.

The Nest Thermostat is a great example, it uses machine learning to learn your preferences for heating and cooling, making sure that the house is the right temperature when you get home from work or when you wake up in the morning.

And More

The use cases described above are just a few of the virtually infinite possibilities, but they’re important because they’re useful applications of machine learning in IoT that are happening right now.

But overall …. We’re Just Scratching the Surface. The billions of sensors and devices that will continue to be connected to the internet in the coming years will generate exponentially more data. Not only will we be able to

predict when machines need maintenance, we’ll be able to predict when we need maintenance too.
Machine learning will be applied to the data from our wearable devices to learn our baseline and determine when our

vitals have become abnormal, calling a doctor or ambulance automatically if necessary.

Beyond individuals, we’ll be able to use that health data at scale to see trends across entire populations, predicting outbreaks of disease and proactively addressing health problems.