In recent posts, we have covered how big the internet of things may be, how much economic value it may create, how network complexity of connected devices increases value, and in this post we will evaluate a similar hierarchy, but instead of network connectivity increasing, here the increase in logic and communication complexity relates to increased value.

The IoT use cases that come to mind for most people, and frequently the examples given, are based upon simple connectivity for passing data from one device to another. Much of the state of current technology emphasis (e.g. Apple’s API or the Eclipse M2M platform) is focused on accomplishing this basic connectivity. However, machines coordinating with more intelligence, more complex language, and with greater and greater scope of coordination will come closer to realizing the true potential.

Let’s investigate the following analogy that illustrates the increasing value of intelligent connectivity for a hypothetical soccer team. At its most basic level, this team is comprised of eleven players on the field, but they have no assigned positions yet. Nobody knows if they are supposed to defend the goal or score a goal. The players pass the ball back and forth but with no real direction. This is basic connectivity. Now let’s assume the players have assigned positions and begin talking to each other—this introduces a level of intelligence into the team such that individuals now have a function and a goal (pun intended). No surprise, the team begins to function a bit better. Now let’s introduce a coach who provides plays for the team to execute. All of a sudden each player knows her role and understands how to collaborate with the rest of the players. The team starts winning games. In order to take this team to the next level let’s assume that the coach resorts to big data tactics. He analyzes the performance of all of his players centrally, compares their performances vs. the competition (his team’s “operating environment”) and uses this intelligence to fine tune both on and off field strategies. Now the team is in a position to win the World Cup. In fact, this is exactly what the current world champion German soccer team did using big data capabilities.

This analogy involved contrived limitations placed upon humans, but in the internet of things world, machines are trying to accomplish the same things and have the same challenges. Basic connectivity and passing data will not be sufficient. The machines will need to develop more intelligence. They will have to understand their role and do more without explicit external instruction to accomplish their tasks and be able to relate to other machines in the context of this task. The machines will also need to develop a more sophisticated vocabulary to provide richer collaboration and be more adaptive and handle novel situations that arise. Regardless of if one machine dramatically increases its capabilities, there is always greater value to be found in seeing the big picture. Analyzing the full scope of the network, leveraging historical information, and being able to evaluate and analyze for continuous innovation will deliver increasing value.

This hierarchy of intelligence, a little more formalized and in an industrial context, increases value through the following four layers:

1) Connectivity – The fundamental requirement is that there is communication between the machines. This is a mature area, but mostly utilizing limited communication protocols, predetermined communication connections, and even proprietary networks. Connecting all the sensors and equipment brings challenges to speak more “languages” across more “communication channels” and interact without predetermined “relationships”. This is where the commercial/consumer IOT segments are providing new technologies that are enabling the more mature industrial machine-to-machine communications.

2) Process on a Device – The current IOT paradigm is predominantly connected sensors with some overall business process. The industrial market, however, is more occupied by machines and making these machines smarter and more autonomous. To do this, the business process must move more and more onto the device itself. An example of this could be a machine that monitors its own health and when needed notifies a maintenance process to schedule repairs instead of sending all performance data all the time. This is known as “Process on a Device”. This is envisioned to be applied to materials and ultimately include the entire life-cycle process. Upon notification of acquisition, a material would know its intended purpose and track its storage, transportation, assembly, use, and even end of life. This provides the ultimate traceability and transparency in the product life-cycle. A product could report on all aspects of carbon footprint, sourcing compliance, and environmental or usage issues impacting quality and safety.

3) Collaborative – Moving more of the business process onto the machine now requires that the machine communicate in higher level ways than merely connecting for data transfer. This requires machines to be flexible in this communication and that the communication be rich enough to support the range of the business processes it is implementing. This could require more complex two-way inquiries from other machines and not merely polling data or notification of conditions. This is not a new challenge to computers trying to execute a business process without human intervention. The history of business-to-business purchasing transactions originated with very rigidly structured electronic data interchange (EDI) communications. The internet helped evolve this to more flexible XML structures, but even this required pre-negotiation of many aspects of the language and terms used in the communication. The current internet is evolving into the “semantic web” enabling users to find, share, and combine information more easily.

4) Big Data Insight – Regardless of the intelligence, autonomy or capabilities of a single machine, there is always more insight to be gained from aggregated analysis of the overall network. Having every machine reporting on every sensor and step in the process is like capturing all the thoughts of all the workers. This will lead to challenges that rival any current definition of big data. Different capabilities for real-time information processing, event and network processing, and new analysis algorithms will provide the final category of data to the internet of things for industrial use.

Just as our unorganized soccer team couldn’t talk, didn’t understand their position assignments, and couldn’t interact with each other grew into a World Cup victor, our internet of connected devices will also mature in what they can do, how they can coordinate, and what can be learned from this coordination. Consider how your product could increase its own value by incorporating more of the process into the device, how it may need to communicate and interact with other products and, lastly, if there are analytical services you can provide to help see the bigger picture.

Thanks again to Chuck Pharris (Twitter: @chuckpharris) for the collaboration.

To report this post you need to login first.

Be the first to leave a comment

You must be Logged on to comment or reply to a post.

Leave a Reply