IoT: A DATA ECOSYSTEM
While definitions abound, IoT may be viewed as a vision of future services which depend on an interconnected data ecosystem. It is so since IoT's implementation and operability depend on a host of technologies and processes which together represent a data value-chain. The term data is prominent since IoT is primarily about the sensor-collection of data, which are then sent via networks to the cloud for analysis. Thereafter IoT becomes about the actionable knowledge that data may produce, for instance, to provide a service, improve operations, and generate revenues. So, in its basic form, an IoT service has the potential to monitor and control any object on the planet, no matter how exciting or mundane it may seem, and to develop knowledge on various levels, including how it is used, where it is used, and what performance it is delivering. Such capabilities potentially present great opportunities for creation of innovative products and services for many markets on a global scale.
Another key item for consideration is that IoT is about automation. Meaning, sensors are primarily connected to the Internet to feed data to cloud services without user intervention. For the purposes of this article, the focus excludes sensors designed mainly for user data-entry via touch screen or standard networked keyboard and mouse.
DATA ECOSYSTEM:THREE PILLARS
To illustrate this data value-chain a bit, it helps to divide its components into three domain pillars:
- The Enablers;
- The Support System; and
- The Business Drivers.
These three areas are born out of a basic "collect → analyze → strategize" data lifecycle thought, which serves to illustrate a transformation of collected data into information, and then to knowledge.
Taking a service provider focus, each pillar is presented in a simplified IoT Data Value-Chain Diagram (see page 1), and high level summary, next.
1. The Enablers
This section reflects on how the data first enter the value-chain, and how then the end-user may interact with the resulting information.
a) Sensors (the data point of entry): The data meant to be collected must enter the ecosystem in some fashion, before they can be analyzed and the information accessed. As a starting point in data collection, this is where sensors play a specific role, whether they are attached to a consumer good (wrist band, shirt, door), or embedded in an industrial product (utility meter, street lights, conveyor belt).
They autonomously collect a specific type of data, which are then fed directly to the cloud without user input. Connection to a short-range network, such as Wi-Fi, is assumed here to enable data movement at the network edge. In some cases, direct connection to a wide area network (cellular, satellite) may be configured, based on a variety of factors.
b) Computing Devices (devices with web-enabled apps): After sensor-collected data are sent to the cloud and analytics performed on them, the option then may exist (depending on industry and application) for the end-user of the sensor product or their representative to interact with that data in a meaningful way. For instance, one may think of their auto mechanic as sort of their representative at service time when the car diagnostic information is accessed from the cloud on an auto shop tablet. Or alternately, when field service engineers use their mobile device to interact with collected machine health data to provide efficient service during a customer call.