The word IoT has meant many different things to different people, none of it is wrong in any sense. It has been just a matter of perspective. Device and sensor manufacturers think of it as the sensors at the center of the IoT ecosystem with some connectivity and software around the sensors to capture and transmit data. Network services providers think of IoT as a secure network that connects a bunch of commodity sensors to a backend data store. BI and Big Data platform providers think of their role in IoT as the most important; after all if you cannot process a large amount of data coming through and extract intelligence from it, what’s the point in putting all these sensors and networks in place.
IoT is an Ecosystem
The best way that I can explain IoT is as an ecosystem, that is a combination of the right kinds of sensors strategically placed, securely connected to a number of different things and transmitting the desired information to a technology that uses this information to make intelligent, value-add decisions leading to well-defined actions.
This notion of an ecosystem is very well explained by a visual that I saw in the IoT Reference Model that was published by Cisco in 2014. While it was built when IoT was still very new and evolving rapidly, I believe that the model has stood the test of time very well.
IoT Reference Model
As you can see from this 7-layer model, it includes almost all the technological aspects of an ecosystem, starting from the physical devices and sensors to the network connectivity to data management and applications and processes. All these layers must come together to provide true end-to-end value for anybody or any enterprise who wants to leverage IoT in any meaningful way. One could argue that there is one thing that’s needed in this Reference Model. We need something that binds these layers and the information flowing through them together in a cohesive and contextual way. This lack of contextualization between these layers has limited the ability to easily achieve true IT-OT integration, and perhaps, also limited as wide an adoption of IoT as technologists anticipated and predicted.
IoT Platform – Enriching an IoT ecosystem
An IoT platform helps manage this multi-layer IoT ecosystem, whether it is managing devices or collecting and moving data or enabling integrations with back-end ERP or other systems. Basically, the IoT platform enables true IT-OT integration. This is an especially important point, because we know that without true IT-OT integration, the value of an IoT ecosystem is diminished. If we cannot have the real-world interact with the cyber-world in real time, cannot exchange relevant, contextualized information and initiate actions, then we are back to the way we did things before the advent of IoT. Some of the key features that a good IoT Platform should have are:
Device interoperability – It is essential that the IoT Platform support multiple communication protocols so that all sorts of devices and sensors can be connected to and communicate with each other. This is the only way we can break down the operational silos within any organization and enable real-time information exchange across the entire process value-chain, as opposed to addressing things within departments and operational units.
Device discovery – In any localized ecosystem (a manufacturing plant, a city, an oil rig, etc.), there will be several devices and sensors installed. If the IoT Platform can discover these sensors and devices, catalog them, configure them and get them ready to transmit then kicking off value-added activity will only become easier.
Information flow modeling – This is key to really enable true IT-OT integration. If we cannot model operational information flow, we will limit the ability of IT processes to interact with physical assets. Only by facilitating innovative ways for business users to interact with physical assets, will an IoT Platform enable them to align their operational objectives with business goals.
Bi-directional information flow – This feature is very important to enable autonomous M2M interactions. Taking data and information from sensors, contextualizing it and applying advance analytics to make decisions is, of course, important. This bi-directional information flow capability is what will take this decision to another sensor and device and initiate automated action.
Application integration – All enterprises have a number of systems that are used to initiate all sorts of operational activities, such as initiating replenishing inventory, opening a support ticket for a repair, or dispatching an ambulance to someone in need. Integrating with these enterprise applications is also very important to automatically and intelligently enable that end-to-end business process.
If we have an IoT Platform that addresses the above key issues or provides these features, and does it in a way that is secure, scalable and easy to implement then we hit a home run. Such a platform makes the ecosystem cohesive and meaningful, but more importantly, it encourages organizations to adopt IoT for generating transformational business value that the promise of Industry4.0 holds.
Finally, you have any other ideas on what should be in an IoT platform, or another perspective to what I said, please do leave a comment or reach out for a discussion.