The Industrial Internet of Things (IIoT) enables increased efficiency and new business models
By networking machines, objects and assets, factories can become smart factories that can ultimately control themselves based on intelligent technologies. The Internet of Things as applied to industry, known as the Industrial Internet of Things (IIoT), is becoming a real business model that is driving industry forward. But what do terms like IIoT and Industry 4.0 actually mean? And what makes them so revolutionary that they will fundamentally alter the industrial landscape, including in Germany?
For us humans it has become natural that we are "always on," meaning continuously connected to the Internet. Our smartphones help us get from A to B and keep us in touch with our colleagues, friends, and family anywhere in the world. We can check the weather forecast, order goods, read e-mails, or buy bus or plane tickets online. There is very little that cannot be accomplished via the Internet using a smartphone or at-home PC.
The Internet of Things (IoT) applies this always-on concept to communication between things: In the IoT, objects are networked with one another or to central computers. But to what end?
To clarify what IoT is and can do, think of a refrigerator that, when connected to the Internet, can purchase whatever it's "out" of directly from the supermarket. The same principle also lets you turn on the heating by smartphone. If you don't want to return home to four cold walls, you can regulate the temperature before your arrival.
IIoT is more of a strategic concept than pure technology
One of the most promising projects based on the integration of things is autonomous driving. It depends largely on the ability of vehicles and vehicle parts to communicate with one another and with central computers. Experts believe it will take another 20 years before truly self-driving vehicles that can safely transport passengers will be on the market.
IIoT, often simply called "Industry 4.0", is more of a strategic concept with far-reaching consequences than just another way to apply IoT. Companies that can consistently rely on the networking of machines, systems, and assets are more productive, as machine runtimes can be optimized and production data used to boost profitability. Operations and maintenance costs drop because machines can be maintained and repaired before they fail. Customer satisfaction eventually rises because needs are met more quickly and on an individual basis.
Networking production: IIoT
From a purely technical point of view, IIoT is the networking of production systems, machines, and assets such as vehicles and devices over the Internet. It uses the same technologies as non-industrial IoT. Such devices include sensors that can perceive or measure their surroundings and send the measurements to a central unit for processing. A sensor can be a device that measures machine oil temperature, an artificial "ear" that listens to noises during machine operation, or a shock sensor that records production system vibration.
All these activities produce data that is sent to the cloud via network connections or processed directly on site (edge). Analyzing these data has direct consequences if they deviate from a "normal" state, e.g., a failure alarm, notification of service personnel of an impending machine failure, or direct machine shutdown in case of overheating.
The data can also be used to manage the business, forecast machine life cycles based on key performance indicators, improve the efficiency of a production method, or achieve better utilization of construction equipment.
Paradigm shift in business models through IIoT: Platform economy
Strategically, the IIoT could make a paradigm shift in company business models possible. To understand why, we need to take a look at what is called the platform economy.
The top five most valuable companies in the world are all technology companies. But something else is critical for this ranking: All of these companies are platform providers. Apple maintains one of the world's largest marketplaces for mobile apps, Amazon is the world's largest commercial platform, Facebook deals in the data of its users. Google, Amazon, and Microsoft are among the world's largest cloud providers, through which they offer companies a platform for new business models.
A major share of their business still consists of selling their own products or posting advertisements. In addition to the sale of physical assets, however, they rely increasingly on mediating between supply and demand, which is referred to as matchmaking. Platforms serve as intermediaries, connecting two or more market participants by means of digital technology.
From the view of corporate users, platforms have a critical advantage: they can be used as a fully-developed infrastructure without the company having to build or maintain it. This is an enormous financial and organizational relief, creating scalable opportunities for new businesses.
Platforms are increasingly becoming the central business model of the digital economy. They expand existing markets and create new ones.
When real assets virtually meet: digital twins
IIoT yields true innovations that would be unthinkable without this technology. This includes so-called digital twins, which virtually replicate real-world processes and machines.
Experts agree that these digital twins will soon be among the most important foundations of digitalization and will create added value that extends far beyond simple integration.
For example, digital twins can be used to digitally replicate aircraft engines or wind turbines to simulate their operation. Failures or signs of wear can thus be detected and addressed before damage or system breakdown occur. The more complex the system, the greater the potential savings offered by digital twins.
The twins operate like real objects but only exist in virtual form. They are therefore much easier to use to simulate new applications or complex, multi-stage business processes that could otherwise only be tested at significant capital expense and risk. Like their genuine counterparts, the digital replicas provide data for predictive analytics purposes, i.e. prognoses and scenarios.
Artificial intelligence – the IIoT booster
The most valuable raw material produced by the Internet of Things is data. When generated properly, data provides us with information and insights into previously unknown worlds. We can understand practically in real time what our machines are doing or when they might fail to function if left unmaintained. They offer valuable information concerning their efficiency and utilization, to allow us to optimize production and minimize downtimes.
The quantity of data grows exponentially in accordance with the number of networked machines, systems, and assets, and soon we will no longer be able to convert these data into actionable information through human skills and abilities. At the same time, computer performance is continuously on the increase: They are always faster and are able to process enormous volumes of data almost in real time. It is primarily these two factors that are driving the artificial intelligence (AI) boom, the beginnings of which we are now experiencing.
AI concerns machines’ ability to "understand" the environment, learn from it, and select a course of action. The method of learning independently from the existing data and generating algorithm for action is called Machine Learning. In the area of IIoT, AI is close to the meaning of machine learning. A currently commonly used method of machine learning using neural networks is Deep Learning.
AI works with algorithms, i.e. multilevel instructions that flexibly respond to conditions that the sensors, for example, report. The algorithm doesn’t care whether the value it receives from the sensor constitutes a temperature, sound pressure, or brightness. However, with the right training, it can interpret the values transmitted by the sensors, identify deviations from the norm, and infer a course of action.
Machines have long been able to outpace humans in quickly and reliably detecting patterns in vast quantities of data. From the moment algorithms interpret the patterns independently and without human guidance, they would have achieved a general intelligence similar to human intelligence, but we haven’t reached that stage yet.
AI, in which humans support before and after training the algorithms, however, already represents an extremely important technology for IIoT, as it can provide new insights into the operation of machines and systems – and also drive production automation toward autonomous manufacturing and factories that can more or less operate themselves. The era of smart factories will only then truly begin.
Previously, automated systems depended on fixed rules: a magnetic strip that shows a robot where to go, or a program that a machine works through step by step. Artificial intelligence will ensure that machines are able to avoid the unknown and unplanned, and – without human prior knowledge, foreshadowing and planning.
IIoT and Security
The cardinal question remains the security of Industry 4.0, an issue that needs revisiting given the comprehensive integration of almost all objects, machines, and assets. Their number alone – market research company Gartner cautiously estimates more than 20 billion networked devices by 2020 – reflects the magnitude of the problem, because every one of these devices is theoretically a contact point for attacks on networks. The task of safeguarding the IIoT and the data produced there is made even more difficult by the variety of devices and network technologies. There is also no simple answer to this challenge.
But experts do agree on one thing. Modern technology for networking and data analysis is not only part of the problem but also part of the solution: in the form of policies that regulate the transmission and exchange of data and that require the use of IT. Encryption mechanisms that protect data from external access during transmission, artificial intelligence that helps with intrusion detection – these are all basic measures that every company should use, embedded in an agile and dynamic security strategy whose effectiveness is continually checked and improved.