In small steps towards Industry 4.0

The route to Industry 4.0 does not have to be a leap or a complete replacement of existing machines for the manufacturing sector. On the contrary: with an approach via small steps, better, more sustainable digitisation results can often be achieved.

The term Industrie 4.0 was first uttered at the Hannover Messe in 2011. Ten years have passed since then. According to a mid-2019 study by IT market research firm IDC on the extent of IoT implementation in the industrial sector, self-optimising systems are still rare exceptions. There are still often impeding barriers between individual parts of production facilities and IIoT is only rarely used as an information hub.

The good news is that, through retrofitting, machines and systems can also be gradually transformed into cyber-physical objects and integrated into the IIoT. Provided the right IT and general control systems are in place, you can equip them with modern, integrated control concepts and increase their level of automation. The cabling and networks of existing systems in so-called ‘brownfield’ environments (as opposed to new systems in the proverbial ‘greenfield’ environments) and the additional sensors required present complex challenges in the field of system automation.

There is no panacea to overcome these, as in each case there has to be alignment with respect to different requirements and individual circumstances. To give industrial machines a makeover and make them suitable for the IIoT - through 'computerisation' - is therefore a logical, future-oriented task of automation companies.

The trick is to use the possibilities at the various points in the overall network without bias. And to adapt them as best as possible to the requirements. When developing a heterogeneous environment, the starting point should be a flexible modular construction kit consisting of autonomous units. In addition, it is good to properly identify the many different technical possibilities. This helps in linking the existing and later added sensors and actuators in the machines with their controls and secondary and higher-level systems.

An agnostic choice of technology and product can avoid over-dependence on the preferred standards of individual hardware manufacturers and ensure a high level of economic efficiency in operation and maintenance. The focus is on connectivity at the various data technology levels.

Organisations also face a huge variety when choosing physical data connections. Within machines, you are more likely to rely on industrial copper cables with M12 connectors, for example. Fibre-optic cables then only deploy in fault-sensitive applications. Because of the often difficult and cost-intensive cabling of existing production facilities, the trend towards external communication is increasingly heading towards WLAN and other wireless transmission networks.

This includes the new 5G standard. Its suitability for data transport in the IIoT is indisputable and it is widely advertised. Numerous manufacturers are also constantly marketing suitable connection devices. However, the actual data transmission will only become clear in practice.

Alternatively, transmission networks developed specifically for IIoT can be adopted. These can allow sensors to communicate directly with IT, bypassing existing control systems. These include the Sigfox global wireless network or the Long Range Wide Area Network (LoRaWAN). LoRaWAN sensors score with particularly low power requirements and very high inherent data security. Therefore, they are ideally suited for integrating remote and even mobile systems.

Additional data and signals from retrofitted sensors are relevant not only for upstream systems, but also for local control, regulation and visualisation systems. Connection to these is usually best achieved via I/O modules. These should be compatible with the existing fieldbus or Industrial Ethernet. Whether DIN rail mounting in an existing or new control cabinet is sufficient influences the choice of components. If this is not the case, the availability of variants with a higher protection class and with a wider temperature range and the possibility of low-voltage power supply for installation directly on the machine frame is essential.

Existing machine controls often lack free capacity. Therefore, manufacturers of IIoT solutions often need to build in a further level of processing. Local computer intelligence takes over the pre-processing and translation of the additional information, communication with the existing control unit and forwarding the data to the outside world. More and more, industrial versions of the flexible, programmable small Raspberry Pi computer are replacing the previously dominant PLC devices.

Whether you use the existing network structure, introduce a wireless network, or connect directly to cloud services (or a combination of these); the switches, routers and protocol converters are central elements of the connectivity upgrade. Like all other components, from cable assembly to surge protection, they do not have to be available only in industrial versions.

IIoT projects are highly individual, makers build components and devices in very small quantities into individual parts. You need someone who can provide the necessary hardware with the right compatibility properties at short notice. Also in small quantities and with long-term availability.