The Industrial Internet of Things

05/15/2013


The technology solutions we create must be easy, flexible, and powerful

Rick Bullotta

Any consideration of applying the concepts from the IoT to the industrial space would be incomplete without addressing the following:

  • Legacy systems and devices—How will they participate in this new architecture, at all levels of the stack? While IPv6 and 6LoWPAN are important moving forward, we need to embrace existing devices and endpoints as well.
  • The IoT and I2oT are not a communications/plumbing problem (or opportunity); they are about creation of useful applications. While standardizing some of the lower level networking is helpful, it will fall far short of truly unlocking the potential and represents only a tiny piece of the requirements. Other critical elements include:

1. A semantic model for discovering, addressing, and consuming the data, services, and events that the elements of the IoT/I2oT will provide. Although the “I” in the IoT stands for Internet, the reality is that the Internet wasn’t necessarily the source of the amazing innovations we’ve seen that have changed our lives. It was in many cases the WWW and related standards and protocols that ran on top of the Internet. The same will be true of the IoT.

2. Highly granular security models that can protect access to very specific device capabilities. This way, we can allow selective sharing and access control, better deal with cyber security implications, and so on.

3. Quality of service (QoS) and security at the network layer. Not all messages and bits that are passed on the IoT and I2oT are of equal importance, and this needs to be designed into the stack. IPV6 offers some capabilities in these areas, but more is required.

Let’s not forget the human side of the discussion. People still represent the sensors, actuators, and knowledge base for a huge amount of industrial processes. Failure to consider how humans will interact in the I2oT will lead to failure!

Despite some vendors’ claims to the contrary, the IoT and I2oT are not simply cloud device architectures. In fact, to be successful, secure, reliable, and capable of performing as required, we need to consider them as a distributed systems architecture. Those of us who come from the industrial automation world have been dealing with these types of problems for decades, and there is much to be learned from past experiences and applied to the IoT and the I2oT. Standards are important, but we need to consider carefully where in the stack to focus our energies first on standardization. For example:

  • Which areas have the most immediate impact/value?
  • How can we address the issue of legacy integration?
  • How can we “future proof” our standardization efforts so that when IPv24 and infinitely fast, zero gravity, powerless wireless communications are available, we aren’t starting from scratch?
  • Consider not only the use cases of the past, but the use cases of the future.

Moreover, how can we embrace some other key elements of the IoT in the I2oT?

  • Location awareness of assets, people, and even data. Data has time, value, quality, and location.
  • Contextualization of data via metatagging and other mechanisms, such as a move from dumb historians to smart historians,
  • Mobile devices and new modalities for interaction, including push-based notifications, search-based access to information, secure connections from anywhere, and so on, and,
  • Extend the concept of the social graph to the equipment, processes, systems, and people in the work environment.

We at ThingWorx are using our extensive experience in the industrial sector (the founders of ThingWorx brought experience from Wonderware, Lighthammer, and Cimnet) to apply those lessons and know-how to the IoT and the I2oT. We share the view that there is huge value to be unlocked. We also passionately believe that the value will be unlocked when we provide technology solutions that are easy, flexible, and powerful. Those elements need not be mutually exclusive. And security and reliability are a given. We also feel strongly that there is much to be gained from sharing experiences and technology in both directions—applying the lessons learned from the open, mobile collaborative, and composable world of the IoT to the industrial space, and leveraging decades of knowledge and experience in delivering reliable, performance driven, distributed systems that exist in the industrial sector.

Rick Bullotta is CTO and co-founder of ThingWorx. 



No comments