IoT, the Internet of Things, has long been established in the consumer sector. Amazon’s Alexa is probably the most prominent example of such devices. Its utility value is constantly increasing. From ordering at Amazon  to a smarthome application with an Internet-capable refrigerator, Alexa can fulfill almost every wish.
This refers to all objects that are connected to the Internet in order to connect them to each other. One goal is to support people and make their lives easier. Ideally in such a way that people don’t even notice. The prime example of IoT is Alexa, Amazon’s Internet-based assistant. IoT collects and stores data on the Internet and is controlled from the Internet. Industrial companies are also interested in making life easier, because in this context, making life easier almost always goes hand in hand with financial savings or optimization of production. The IoT applications for industry are, quite conclusively, referred as IIoT, Industrial Internet of Things. In industry, the quest for Industry 4.0 is becoming stronger and stronger. Industry 4.0 means that machines are networked with one another in order to advance automation and at the same time improve production.  56% of the companies asked by the weekly “Computerwoche” in the course of the study “Internet of Things 2019/20,” stated that they consider IoT to be highly relevant. The Smart Factory is the goal that many companies want to achieve. However, a factory can only be smart if one “head” knows what the many subsystems are doing. So the path inevitably leads directly to IIoT, the Industrial Internet of Things.

The added value

However, there is still the question of why the effort should be spent. As an example we look at producing machines. Machines have a life cycle. And depending on the purpose and location of use, it can be longer or shorter. If these machines are now equipped with sensor technology, it is possible to monitor various parameters, such as effective torques, temperatures or vibrations during the entire life of the machine. These data can then be evaluated. One possible use is the comparison of machine data shortly before failures, which is called “predictive maintenance”. If patterns appear, more attention can be paid to when these patterns will reappear. Pattern recognition can be based on AI, artificial intelligence or machine learning. This is particularly relevant for manufacturing companies, since failed production machines can result in significant loss of earnings and problems with customers. Digital Twins represent another potential use. Digital Twins are extremely precise, virtual copies of the machines in operation. Via IIoT interfaces, data from the operation of the machine can be transferred to the virtual twin. Digital Twins can then be used to simulate and test different machine stages. Starting with pre-series tests, through to optimizing the machines during operation, almost anything is possible.

Storage options

Machines are often equipped with sensors for several parameters. Whether it are vibration measuring devices, sensors for temperature measurement or other sensors that measure, for example, the acting torques in machines. They all record enormous amounts of data. Of course, there are different ways to store them. However, the storage in a decentralized system has the obvious disadvantage of decentralization. In order to access all data, more or less many storage locations must be accessible.

To provide centralized storage, for example in a cloud or on a server, the data must first be collected at the machine and then transferred. IIoT modules can take over this function.

Opinions differ as to which of the two options is the path to success.

It is a fact that industry often decides against storing data exclusively in an edge storage device. Exclusive edge storage is certainly not a viable solution in view of the different company locations. The most common middle course: a cloud in the company’s own intranet. This allows all relevant persons to access the data records in a protected area.

Security

Data protection and data security are topics that concern private consumers as much as companies. Starting with personal data, up to the recorded data of operating machines. The study “Internet of Things 2019/20” by “Computerwoche” shows that data protection and security are the main reasons why most companies do not invest in IoT. Internal clouds are one way of collecting and securely storing data. Without a direct Internet connection, securing is much easier. Servers are physically easy to protect, and by integrating them into the intranet, access can still be granted to all people in the company who need to work with the data.

Another point that is not necessarily associated with security at first glance is the dependence on other companies. If a company now outsources all its data to external service providers who offer data center services, then its own company is dependent on this same service provider in a particular way.  At the same time, outsourcing can also have the advantage that specialized personnel will take care of the security of company data. The own IT specialists are thus available for more productive activities than creating backups or installing patches and updates.

An often underestimated topic in IT security issues is the physical security of servers. In addition to fire protection, water protection, dust protection and electronic corrosion, external access must also be taken into focus. The core questions here are who could potentially have access to the server cabinets, how the server cabinets are secured and whether server cabinet openings can be logged. If logging is possible, then it will make it easier to clarify unauthorized access in the event of an incident.

The Internet of Things offers countless possibilities for implementation in a professional environment. In addition to the use of machine learning or AI to prevent production downtime, IIoT applications can also be used to optimise production. IoT also enables the security level to be increased when doors are fitted with door contacts, so that door openings can then be forwarded to a control centre via the Internet. Because of the many possibilities offered by IoT integration, a concrete plan must always be considered on a case-by-case basis.  There is no generally valid statement about the right or wrong approach to issues that can be approached in so many different ways. Every company must find the best solution for itself.