The Internet of Things Requires Smart Sensors and Smart People

Stephen DeAngelis

April 21, 2014

Nearly everybody has heard about the Internet and the World Wide Web. They have helped connect the world and have spawned entirely new industries. Not as many people have heard about the Internet of Things (IoT — sometimes called the Internet of Everything or the Industrial Internet). Whereas the Internet primarily connects people (P2P), the Internet of Things primarily connects machines (M2M). Kathy Pretz explains, “Sensors, actuators, and RFID tags have been around for a couple of decades. The identification and tracking they make possible — noting what an object is and where it has been — are being used to manage inventory, monitor machinery, and track packages and livestock, to name a few examples. Minuscule sensors are just about everywhere, including automobiles, cellphones, clothing, credit cards, exercise equipment, gaming consoles, and along highways. Now the next technological phase is being ushered in: the Internet of Things, a network of objects made possible by the Internet, as well as by Wi-Fi, tablets, smartphones, and apps. Today’s mobile devices are outfitted with a host of sensors, including accelerometers, gyroscopes, and microphones, to say nothing of compasses, GPS capability, and cameras — all sharing data wirelessly over the Internet.” [“Smarter Sensors,” The Institute, 14 March 2014] Ericsson, the Swedish technology firm, envisions that by 2020 the IoT will include over 50 billion devices that are continuously connected and communicating. Pretz reports that some analysts believe that Ericsson has shot way below the mark. She indicates that some analysts believe that “the number of connected machines and devices will grow to 1 trillion by 2022.” As Pretz noted above, RFID tags are among the sensors that will help connect devices, which is why that market is expected to grow. “Research conducted by IDTechEx, and published in RFID Forecasts, Players and Opportunities 2014-2024, finds that the RFID market – including tags, readers, software and services for passive and active RFID – will grow from $7.88bn in 2013 to $9.2bn in 2014. IDTechEx expects that the RFID market will reach $30.2bn in 2024.” [“RFID Market to Grow to $9.2Bn This Year, Surpass $30Bn by 2024, Report Finds,” by IDTechEx, SupplyChainBrain, 25 March 2014]

Pretz lists some of some the current connections being made over the IoT which include: “controlling home electronics from the office, locating empty spaces in parking lots, [and] checking carbon monoxide levels.” But it won’t be consumers who are the primary users of the IoT; it will be industry. That’s why General Electric likes to refer to the IoT as the Industrial Internet. Pretz reports, “The IoT is expected to affect how businesses operate, including unlocking new revenue from existing products and inspiring new processes. An Economist survey of more than 770 businesses around the world found that 75 percent of them are already exploring the IoT and 95 percent expect to be using IoT applications by 2016.” Mark van Rijmenam believes the IoT will usher in a revolution in the manufacturing sector; but, before it can, it has to overcome several challenges. [“The Industrial Internet Will Bring a Revolution to the Manufacturing Industry,” SmartData Collective, 16 October 2013] He explains:

“First of all, turning a factory into a smart factory requires a large investment and a new way of working. It will also require big data standards and an ecosystem to ensure smooth operation between different companies, which has yet to be developed. Of course, the vast amount of sensors in the machinery will create a lot of data, … probably driving us into the brontobytes era pretty soon. This will require powerful analytics that can handle that vast amount of data. In addition, security of all the data becomes a vital issue for continuation of the factory as malware in the Industrial Internet could trigger direct physical destruction instead of ‘just’ affecting sensitive information.”

Bruce Schneier agrees with van Rijmenam. “We’re at a crisis point now with regard to the security of embedded systems,” he writes, “where computing is embedded into the hardware itself — as with the Internet of Things. These embedded computers are riddled with vulnerabilities, and there’s no good way to patch them. … If we don’t solve this soon, we’re in for a security disaster as hackers figure out that it’s easier to hack routers than computers.” [“The Internet of Things Is Wildly Insecure — And Often Unpatchable,” Wired, 6 January 2014] Pretz adds another challenge to the list: people. She reports, “Businesses cite a lack of employee skills and knowledge as the No. 1 obstacle to their use of the IoT, according to the Economist report.” The only way to tackle this last challenge is by teaching people the skills that are needed to administer the IoT and work on the devices that will be connected to it. And it’s not too soon to start training the next generation of employees. That’s why I, along with a few colleagues, founded a non-profit organization called The Project for STEM Competitiveness. If America is going to remain competitive in the global marketplace, we need employees equipped with the right skill sets.

The Internet of Things is not a vision of some future system. According to Cisco, the IoT helped “generate at least $613 billion in global profits in 2013. [“Internet of Everything Expected to Generate $613 Billion in Profits in 2013,” M2M World News, 24 June 2013] Larry Dignan reports, “Machine to machine connections, also known as the Internet of Things, sit in the intersection of cloud computing, big data and mobility and could be a game changer for many industries.” [“M2M, Internet of Things: 10 key points,” ZDNet, 26 June 2013] As the headline of his article proclaims, Dignan gleaned “10 key points” from “a panel of executives gathered by ZDNet, TechRepublic and CBS Interactive B2B’s custom content group” last year. The first point deals with the size of M2M and the panel of experts disagreed with the predictions of the experts cited above. Dignan writes, “IDC analyst Carrie MacGillivray said that projections of 50 billion connected devices by 2020 may be a bit ‘aspirational.’ If you carve out sensors the total is more like 100 million. ‘M2M is just at the starting gate,’ she said. There will be a lot of growth over the next decade, but an ecosystems of vendors, partners and standards need to be built.’ The second key point raised by the panel involved compliance.

Compliance, regulation and standards matter. Sanjay Ramaswamy, vice president of planning and business systems at Johnson & Johnson, said M2M has a lot of potential for healthcare, medical devices and pharmaceuticals. The catch? ‘We need to see how the ecosystem shakes out,’ he said. Specifically, security standards will matter for regulated industries that face numerous compliance rules. Sensors will also need to be standardized so they can talk to each other and share data. Ursuline Foley, CIO at insurance firm XL Capital, said that privacy laws will also be important around the world.”

With analysts like Schneier and van Rijmenam worried about IoT vulnerabilities, you can be assured that compliance issues will remain at the top of the IoT priority list for a long time. The next key point discussed by Dignan involves consumerization.

Consumerization and M2M. Gilad Lotan, chief scientist at Betaworks, a company incubator, said that the M2M movement could be driven by consumers. For instance, people are measuring health and statistics via devices such as Fit Bit and Nike’s Fuel Band. These early adopters could in some way become de facto sensors and tie into health care outcomes.”

Most analysts believe that industrial M2M will be exponentially larger than P2M connectivity. There are simply a lot more machines in the world than people and not all people can afford to buy and use wearable devices. For further discussion about wearable devices, read my post entitled “Wearable Devices and the Quantified Self.” Emerging markets are the subject of the next point raised by the panel of experts.

Hurdles in emerging markets. Unicef’s Stefan Zutt, CIO, said that emerging markets often have technology ceilings that hamper the ability for sensors to interact. As a result, inventory and asset tracking has to be a bit more manual than hoped. He also wondered whether government or the private sector could truly drive an M2M strategy.”

Although infrastructure is a struggle for most developing nations, small countries like Rwanda are demonstrating that, with enough commitment, challenges can be overcome. The fact that most developing countries leap-frogged landline telecommunications in favor of mobile telecommunications also demonstrates that consumers in developing countries are not technology averse. As far as concern over who is going to drive M2M strategy, I wouldn’t worry. Like the Internet before it, the IoT will develop just fine as users determine what it is they need to do with it. The next point raised by Dignan discusses necessary steps that companies must take if they want to fully realize the potential of the IoT.

Necessary precursors to M2M. XL’s Foley said that there’s a line to walk between getting a company’s data house in order via standard definitions and pursuing M2M. Companies need to do the former in a way that doesn’t put off M2M strategies too far in the future.”

Whether the IoT involves 100 million, 50 billion, or a trillion devices, a lot of data is going to be generated by the IoT. Foley is correct that companies need to understand how they are going to deal with all of that data before they are drowning in it. Looking to the future, the panel next considered which industries are most likely to adopt M2M strategies first.

Verticals best suited for M2M. Verizon’s John Sullivan, director of M2M platform for the telco giant’s enterprise unit, said transportation, healthcare and utilities are key areas for M2M demand.”

Supply analysts have been calling for more visibility in the supply chain for years. They will welcome better connectivity in the transportation industry. Such connectivity could provide the backbone for even greater supply chain visibility. The panel next discussed innovation and M2M communications.

Creativity and M2M. Peter Goldey, CIO and data scientist at OnboardInformatics, said the real hotbed for sensor data is likely to appear where we least expect it. The key for the industry is to create an ecosystem so developers can take advantage of interoperability to network and analyze various sensors and data points.”

It has always been the case that users surprise inventors with the creative things they do with new technologies. Just look at the amazing ways that people have used Microsoft’s Kinect technology. I’m certain the same thing will occur as the IoT matures. Dignan next discusses vendors.

Go to vendors. Ramaswamy said that M2M vendors are likely to be existing IT suppliers who have acquired smaller innovative companies. ‘Traditional vendors and consulting services understand the legacy stuff we have and can be realistic,’ he said. ‘We don’t have to be on the bleeding edge.’ Smaller companies can be players too, but will have to prove they can integrate, know security and can provide analytics, noted Goldey.”

Big companies do seem to be in a buying mood; especially for companies that have anything to do with Big Data. Getting back to my point about the importance of having people equipped with the right skills, Dignan indicates that the panel of experts stressed that properly trained people will be in high demand.

CXOs and M2M. The panel was unanimous that M2M and business processes will become one. As a result, business line leaders will be driving the M2M agenda with heavy involvement from IT. Everyone also agreed that people who can meld engineering, statistics and business understanding will be in demand from a career perspective.”

The last “key item” discussed by the panel of experts and reported by Dignan involves the missing ingredient that will really make the IoT take off.

What’s missing? Analytics tools that can enable the masses to make sense of M2M data was a missing ingredient, noted MacGillivray. Zutt also wanted more data and forecasting tools embedded in sensors. Business models for M2M were also tricky since traditional carrier billing approaches like monthly charges don’t quite work for sensor information.”

Can there be any doubt remaining that we have entered the Era of Big Data? As I have stressed in the past, it’s not the amount of data you have but what you do with it that matters. Tomorrow’s winning companies won’t be those with the most data they will be the companies that use their data to the greatest effect. That means they must master Big Data analytics. Because the amount of data generated by the IoT will be so unimaginably large, I predict that cognitive computing systems will be tasked to help manage the workload. The task will simply be too large and too complicated for current computing systems to handle. The “smart world” we are heading into will require smart systems and smart people to achieve the dream.