Earlier this week, we saw how TechCrunch attempted to model the Internet of Things ecosystem. In this post I propose an alternative model, albeit one quite similar in structure . I will not attempt to reproduce in comparable detail the vast number of companies and organisations involved, but will simply provide a representation on how the overall model can be represented. In common with many other models, I too split the connected things ecosystem into industry verticals and horizontal enablers. This means that there are essentially two types of players in this field: those companies who provide the technology, services, infrastructure and other capability to allow a company create a ‘smart’ experience; and new or established companies who are using these enablers to create new products or enhance in some way existing products or operations. Let’s look at these two categories in a bit more detail.
Industry Verticals
The industry verticals comprise all services and products offered to an end-user, and the whole swathe of industries that can benefit from IoT technologies. I have separated the verticals into two distinct sections, the Personal Domain and the Public and Industrial Domain.
The Personal Domain includes all the products and services that are ‘owned’ by and end user, including anything within the home, the car and any wearable or portable devices. What all these have in common is that the data and information generated by any of these devices and services is personal data that belongs to the end user, and as as such must be kept secure and managed on his or her behalf. Within the Personal Domain, there are a number of devices that are associated with a single user and are often worn or carried. This is the Connected Self category and includes fitness and health monitors, smart clothing and smartphones. The second category includes all those systems found in the home and car, which may be shared by the whole family or household. Home automation, heating and cooling, home appliances, entertainment systems, connected cars, security and monitoring all fall into this category.
The Public and Industrial domain is divided into Smart Cities and Industrial applications. I have used a broad definition of Smart Cities, which I have taken to contain all public spaces (such as retailers, shopping malls and transportation hubs) and public services (such as mass transportation, power systems, urban traffic systems, waste management) with which individuals interact. For example, micro-location technology such as beacons may be used in shops to provide appropriate advertising, while motorists may be directed towards an available parking space. As for Industrial applications, this includes all industries where a combination of sensors, smart devices, connectivity and analytics are used to drive product or process improvement. This can be anything from predictive maintenance of lifts, to smart logistics systems and automated agriculture applications. There is no meaningful interaction with the end user for any of the scenarios in this category.
Horizontal Enablers
As discussed, the horizontal enablers comprise all those technology and service elements required to make a smart connected system. A key aspect of this model is that it requires all components to be implemented to be considered a true ‘smart’ or Internet of Things system. For example, a home automation system that does not connect to the Internet or does not act automatically upon the data generated cannot truly be considered as being a ‘smart’ system. The same applies for industrial applications. For the purpose of this model, smart systems are those that connect physical devices to centralised computing centres (usually cloud, though not necessarily for enterprise or industrial applications) and act upon the data generated. Therefore the key building blocks are:
- The sensors and devices capturing what’s happening in the physical world, and often interacting with it. These can be fitness bands with heart monitors and motion sensors, proximity sensors in roads, but also more sophisticated objects such as drones and robots.
- All devices need to communicate back to a platform, either directly, or through a gateway, using a plethora of wireless, cellular and fixed communication networks
- A critical element is the Security Infrastructure. Although it is shown as a standalone element, it needs to be built into each component of the system, including the devices, apps, gateways, hubs, communications, servers and cloud platforms. This is the element that prevents personal and enterprise data from being lost or compromised, and ensures the integrity and trust in the system.
- I have included a layer called IoT Platforms. This is intended to include all the middleware and software that allows new devices and sensors to connect to the network, be discovered, and communicate. There are a number of standards and methods being proposed, including the web of things, AllJoyn, and the Open Interconnect Consortium
- As discussed already, Data Analytics is fundamentally what makes an IoT application a ‘smart’ system – and this is where the tools of big data and predictive analytics are employed
- All the above is underpinned by the IT infrastructure, be it cloud-based, hosted or hybrid, which often integrates into legacy IT and business systems.
What is driving all this change?
So why is all this technology coming together now, and why does the “Internet of Things” appear to be at a peak of hype? There are two ways to answer this question. From a technology point of view, for the first time we are seeing a coming together of all the building blocks that make it feasible to produce distributed smart systems of things. Low-cost, high-power cloud computing is enabling advanced analytics and processing of vast amounts of data. The barrier to entry for an innovative start-up to create a sophisticated cloud-based business is now very low, which is in turn providing a competitive threat to established businesses forcing them to revisit how they provide services. The smartphone revolution has driven down cost and power consumption of high-performing sensors and processors, which has given rise to wearables, and all this is being brought together by changes in communication networks, including low power networks, faster cellular networks, and changes in network architectures.
All this technology would be meaningless if there wasn’t any business value to be gained. A previous post (which will be updated) has already looked at the business drivers. But put simply, all this technology can generate real business value. In the shortest term, the clearest value will be produced by optimising existing processes, reducing costs by making use of better insights (e.g. reducing maintenance costs, optimising logistics etc). But there exist a whole suite of other business motivators, ranging from the car and home insurance companies who want to use connected devices as a means of gaining a better risk profile of their customers, which also providing greater opportunity to differentiate their services to the Internet giants who want to extend their lucrative advertising-based business models beyond the virtual world to the physical world through micro-location, beacons etc.
In a nutshell, this model is only one of a very large range of ways in which the Internet of Things may be characterised. I will be addressing my attention to the individual blocks and elements of this model in subsequent posts.