The technology industry is full of buzzwords and trends and the Internet of Things (IoT) can seem like just another one of those. However, IoT is much more than wi-fi light bulbs and fitness trackers. Consumer products are becoming connected but it is worth looking at the impact IoT is having or will have across many other industry verticals.
What is IoT?
The term refers to any device that is built or augmented to send and receive digital data, where that data then is analysed and acted upon. IoT tends to refer to the interface between the physical world and a digital one. There is no single technical architecture, pattern or application that defines IoT. Typically, IoT systems have some degree of distributed physical sensing with connectivity across standard ethernet of Wi-fi or some of the newer radiofrequency approaches, some open source some proprietary, such as LoRaWAN, NB-IoT, Sigfox. The data is transferred using standard protocols, such as MQTT and routed to the appropriate places with tools from an IoT platform (there are around 700 of these in various forms). Initial IoT deployments focussed on taking all the data and pushing it to a cloud platform for processing, but more distributed architectures are emerging processing data more locally at what is called the Edge. This is also further evolving to fully distributed computing with execution venues ranging from the endpoint devices themselves all the way to the cloud and anywhere in between. This is sometimes called Fog computing.
The extra I at the front of IoT stands for Industrial. The term is used to describe large-scale manufacturing systems in discrete manufacture, continuous process and energy and utilities. It is here that we see an intersection between the IT industry and what is often described as OT (Operational Technology), the factory machinery. Industrial systems have been connected to one another and collecting data for many years. Programmable Logic Controllers (PLCs) and other elements within Supervisory Control and Data Acquisition (SCADA) have provided industrial automation since the 1970s. Data from production processes is typically stored in a long-term Historian database. Many of these manufacturing systems are based on custom embedded code, tweaked and adjusted per machine over the years. Data is often captured by not used unless it is needed to investigate a problem after it has happened. The OT world is now able to leverage the developments from IT, with both large-scale virtualized systems, such as cloud, and advances in Machine Learning and Artificial Intelligence to make sense of the machine data closer to real-time. The existing historical data provides a rich contextual training set for these AI/ML tools too. The upshot is that a much greater efficiency can be achieved across industrial processes because problems can be spotted and adjusted sooner, in minutes rather than days or weeks. It is possible to engage in predictive and condition-based maintenance of machinery and to deliver more timely and accurate information to front-line workers, in some cases using Augmented Reality systems.
As a Service
IoT forms part of the digital transformation of the enterprise. As shown above, IIoT production lines can be optimized but the IoT doesn’t have to stop there. Products that are created and shipped can also be instrumented and deliver data back to the manufacturer. This direct connection with a product has started to create the potential for everything to be delivered as a service, and for manufacturers to be able to offer over the air upgrades to previously isolated and disconnected products. An example of this is Tesla cars receiving new features such as improved suspension settings in an overnight software patch. This also applies to the machinery being used in the factory, suppliers can offer meaningful service contracts to help their customers get the most from the equipment.
In many ways IoT is a unifying approach, it offers a way to describe the coming together of many systems and crosses previously separate silos. Of course, crossing such boundaries opens up many more organizational and cultural questions than the simple technical ones that IoT solves.