Data Science

How Sensors Will Shape Big Data and the Changing Economy

As we enter 2015, the long hyped Internet of Things is starting to take shape. We’re starting to see intelligent devices, wearables, and connected items that aim to make our lives easier.

The name however – the Internet of Things – slightly obscures the underlying nature of this new system.

In 2012, Vint Cerf, Chief Internet Evangelist at Google and “Father of the Internet,” clarified what the emerging IoT actually is: the internet of sensors and actuators.

And to clarify more, all of those sensors will be throwing off data. So, the IoT is an ecosystem of sensors, data-flows, analytics, and actuators. And this ecosystem of sensors, data, and computation will have a dramatic impact on our economy.

A trillion sensor world

As we talk about smart phones, tablets, watches, and connected objects, we need to understand that those “things” will be packed full of sensors.

To put some numbers on this, Gartner forecasts 25 billion connected things by 2020, a 566% increase over 2014. And, this is just the number of “objects.” The number of individual sensors could top one trillion as soon as the early 2020s according to estimates from several companies and industry groups.

This increase in sensors is being driven by falling sensor costs and new fabrication techniques. For example, the startup company mCube (a maker of microelectromechanical systems sensors, or MEMS) is using a new fabrication method to create motion sensors that are “smaller than a grain of sand.” These are said to be “the smallest accelerometers in the world.” They are now small enough and cheap enough, that mCube envisions a world where motion sensors are added to “virtually everything that moves.” mCube’s vision is a world where all moving things are ubiquitously instrumented, collecting data.

We can already see this vision of a ubiquitously instrumented world becoming reality. For instance, this year, a wide range of wearables are incorporating sensors.

The most obvious sensor-containing products are watches, like the Apple Watch and Samsung Simband. They not only contain motion sensors, but are also packed with other types of sensors that measure biometrics such as heart rate, temperature, and heart rate variability.

Moreover, the Samsung Simband not only has built-in sensors, but also room to accommodate additional sensors from third parties. So as new sensors are created, they can be “swapped in” and integrated into the device.

Of course, all of these sensors will be throwing off data, so the Apple Watch and Simband devices are being integrated into development platforms that record that data and make it available to developers.

Apple is creating HealthKit and Samsung is creating the SAMI development platform. Essentially, these development platforms gather data from the watches and other instrumented objects of the IoT. This data is then made available to developers for the creation of new applications and analyses.

The Samsung SAMI platform and Apple HealthKit are simply small-scale ecosystems that mirror the much larger IoT ecosystem: a vast network of sensors, data, and analytics.

Transforming the planet into vast streams of data

As we instrument larger numbers of objects, all of these sensors will be creating data streams. As we build out this network of sensors, we will be generating vast quantities of new data that need to be analyzed. In a way, these sensors are really the front-end for a vast data system. The sensors will instrument the planet, turning larger swaths of the physical world into data-streams. And the data is where a large amount of the value will reside in the IoT.

McKinsey and Company noted that in order to capture the full value of connected devices and the Internet of Things, organizations will need to have the infrastructure and resources to “make sense of the flood of data” produced by these connected devices. Building the IoT and back-end data processing systems will require deep investment in analytics tools and talent.

As much as “big data” has been the subject of significant amounts of hype, it is real. In fact, as we create this trillion sensor world, big data may become “astronomical.”

And that data will be very, very valuable.

The trillion dollar Internet of Things

Cisco estimates that the IoT could have an economic impact of $14 trillion by the early 2020s. The data generated by these sensors could impact almost every major industry, including healthcare, transportation, and energy.

As this array of a billion devices and trillions of sensors comes into being, skilled data scientists and developers will need to build out the underlying infrastructure. We will need to develop the systems that analyze these astronomical quantities of data and turn it into valued insight and action.



How Sensors Will Shape Big Data and the Changing Economy Joshua EbnerJoshua Ebner is the founder of Sharp Sight Labs, a company that teaches data science to individuals and helps companies optimize their data strategy.

He has a degree in physics from Cornell University and has worked as a data scientist for several Fortune 500 companies, including Apple and Bank of America.


Image Credit: Eric Fischer

Previous post

Making the Most of the Internet of Things

Next post

DataHero Pairs with Marketo to Help Marketing Automation Customers Tap Analytics