The Other IoT

What happens in the home, in the car, or in the tiny electronic devices people carry around with them or wear on their wrists or implant in their bodies is the focus of marketing by thousands of companies all over the globe. After all, the Internet of Things, in some shape or form, is widely expected to the “next big thing,” or at least provide the foundation for many next big things.

Far less glamorous is what is happening in industry, but it’s there that the has really taken root and paid dividends. Rather than hype with uncertain future, the so-called Industrial Internet of Things, or IIoT, is a business investment with well-documented returns. But those returns are so company-specific, or industrial segment-specific, they generally are ignored by analysts and media. In fact, there is no way to know exactly how large the opportunity is other than by collecting data from thousands of companies, and there is no reason why anyone would be incented to do that because there is no consistent way to quantify it or benefit from it.

But there are some elements of the IIoT where it does impact more than one company—generally in the realm of standards-based communications. And it is there that the first glimpses are occurring for just how widespread this technology adoption has become.

Industry 4.0
One area of particular interest is a methodology called “Industry 4.0,” which has gained attention primarily in Germany. If adopted, it could be the most significant shift in manufacturing since the creation of six-sigma quality control at Motorola in 1986 and the lean manufacturing system developed by Toyota starting in the late 1940s.

“The key change is that it gives power to the object being manufactured, not the machine,” said Laurent Dardé, marketing director for security and connectivity at NXP. “The object knows what has to be done, not the machine that makes it. So there is data for preconfiguration.”

Manufacturers have been experimenting with this kind of technique since the 1990s. When IBM was still in the PC business, it introduced technology that would let companies burn their corporate image—applications, security schemes and networking connections—into new PCs sitting on their loading dock rather than individually loading those applications and features onto each computer after booting up. In addition, chipmakers have been creating superchips for just as long, allowing them to remotely turn on features for some customers and different features for others. The new wrinkle is building that kind of intelligence into the device in the first place, rather than externally programming firmware or embedded software.

“A good example of where this can be used is with a DECT phone,” said Dardé. “The hardware is the same all over, but each country has a different protocol. With smart objects, when you sell that phone you can configure it for the region. It adds flexibility into the warehouse.”

But that’s only scratching the surface of where this kind of technology can be used. Understanding manufacturing from the product outward, as opposed to the machine inward, can help improve quality and identify problems in the manufacturing process. Moreover, the built-in communication that the IIoT provides, if standardized, can help improve serviceability in the field.

“A good example of this is when you have a washing machine full of water with clothes inside and it breaks down. You phone the call center and they ask what the model number is, what the serial number is. At least they’ve standardized the place where the serial number is written, but if it’s full of water you don’t want to open the door. But if you can read that information with a phone and send the information into the cloud, the diagnostics can be done in the cloud. They can do a reset or software update or send out a guy with the right spare part. The process is quicker and better for the end user and cheaper for the manufacturer,” said Dardé.

The same can happen on assembly lines and with hard-to-service technology. Sending out a person or team to fix a problem can be half the cost of determining what the problem is first, then sending out that same repair person or team to actually do something about it.

Standard protocols
While this all makes logical sense on a grand scale, the big challenge comes from standardizing the communications.

“The problem is that the Industrial Internet of Things has grown up independently as its own silo,” said Tony Massimini, chief of technology at Semico Research. “They have their own communications protocols for factory automation and building controls. The new challenge is putting it on the Internet. That means getting the right connections, using wireless protocols such as ZigBee or WiFi, and adding security.”

In many ways, the IIoT is a legacy system, created because it made sense from a return on investment standpoint. But at this point, the equipment is dated.

“The IIoT is largely the rebirth of M2M communications,” noted Bernard Murphy, chief technology officer at Atrenta. “It’s been around for awhile, but there has been no necessity to tie it together. And there is a lot of infrastructure already in place that will have to be updated or changed over.”

One of the big issues companies are wrestling with, however, is that while it makes sense to connect everything together, there are inherent risks in doing that on the security side.

“The challenge is to protect the companies once they’re connected, and there is a major effort under way for that,” said Massimini. “The big advantage is that compared to the consumer space, companies can afford to put in more security. That includes everything from hardware encryption/decryption, secure boot and memory. There are about 28 companies doing chips for that. In some cases these are low-cost microcontrollers where they add hardware features, which are recognized as more secure than straight software encryption. But they do add cost. And they have to be low power and be able to run for 10 to 20 years, which means you need to be able to upgrade the firmware.”

Not all of this has to be done from scratch, though. Many of the developments underway in the automotive industry involve the same companies that provide technology to industry. In the 1980s Bosch developed the Controller Area Network bus standard, for example, a message-based protocol that allows microcontrollers to communicate with each other without a host computer sitting in the middle. The technology originally was aimed at the automotive market, but it has been widely adopted by the industrial sector, as well.

“If it can survive in a car, it also can survive in a factory,” Massimini said. “But the original protocol had zero security. No one ever thought you would connect anything to a car.”