Intelligent factory

Andrea Gillhuber,

Data-driven production

When two technological worlds collide, this usually creates friction, but in the best-case scenario it can also lead to new solutions that are significantly more than the sum of their parts. This opportunity currently exists in the Industrial Internet of Things (IIoT). It is important to generate information from operational data that can be used for the entire business. This is entirely possible with the right tools, as this example shows.

New solutions in the industrial Internet of Things. © Crate.io

The technical development is characterized by the convergence of previously separate applications, which also required a completely new organizational orientation. Two decades ago, IT and telecommunications were still completely autonomous areas with their own ecosystems and specialist departments, but the convergence has opened up new ways of communicating. Today, telephony is an integral part of corporate networks. Even if the benefits such as joint administration of access rights, new functionalities and open interfaces are now taken for granted, the process has taken a number of years.

The situation is similar in manufacturing today. OT, or operational technology, still leads a life of its own in many manufacturing companies. Even though the physical equipment in the factory, such as numerically controlled machines, robots, sensors, etc., is now largely digitized, networking is rather limited. Proprietary protocols are still commonplace and responsibility lies in separate departments.

Advertisement

In IT, or information technology, networked data-centric computing based on open interfaces is the rule. It creates the basis for processing, storing, exchanging and securing all forms of electronic data with the aim of intelligent corporate management.

Improving effectiveness

Consequently, the task today is to capture, analyze and store the enormous amounts of data from production and make it available intelligently for subsequent applications. What's more, it is important to derive immediate actions from the information obtained in order to ensure the continuity of the production process. The best way to do this is not only by quickly solving any problems that arise, but also by using predictive functions to prevent them from occurring in the first place.

The challenge for all manufacturing companies is to increase the effectiveness of the entire plant. Key figures such as NEE (Net Equipment Effectiveness) or OEE (Overall Equipment Effectiveness) are important management tools. Practical concepts are required to improve these. Important measures include reducing waste, increasing the speed of production lines, faster changeovers and avoiding downtime. These goals are countered by the increasing complexity of production facilities combined with a lack of trained specialist personnel.

OT meets IT

The solution lies in the implementation of data-driven production with the aim of enabling actions directly on the production line based on real-time data, relieving production staff of mundane monitoring tasks, shortening operator response times and providing production staff with tools to improve the entire process.

As logical as this sounds, it is difficult to implement in practice. It is characterized by the coexistence of different systems, the limited interaction between IT and production, non-networked systems and devices as well as fractional responsibilities of departments. What's more, the collection and management of the enormous amounts of data alone is anything but trivial. Machines, sensors and devices in production continuously generate data every millisecond and in a wide variety of formats. However, suitable databases that are able to process unlimited amounts of data in a wide variety of formats in real time are rare. Although classic SQL databases offer the advantage of convenient programming and the availability of trained specialists, they suffer from the necessary performance. Specialized NoSQL databases are better suited in terms of performance, but are proprietary and therefore not very integrative. Many companies therefore find that several databases coexist, which is associated with redundant administration, overheads, data inconsistency and other difficulties.

Optimization of the production process with notes and reports in real time. © Crate.io

A suitable database is a basic prerequisite for the implementation of data-driven production. It must be able to record and analyze complex time series in a wide variety of formats with high scalability, store them securely and make them available for company applications in real time. But that alone is not enough. It is much more important to develop an infrastructure on this basis that is capable of improving the entire process, both in terms of functionality and cost-effectiveness.

Manufacturing companies have a core competence in the production and marketing of their products. Although functional IT is a basic requirement, it is not necessarily an element of a strategy that requires considerable investment in infrastructure and personnel. In fact, cloud solutions today offer the possibility of enabling even complex processes without having to invest in in-house IT infrastructures. This means that costs can be calculated and availability is guaranteed.

Example: Alpla, manufacturer of plastic packaging

The example of Alpla shows what a solution can look like against the backdrop of these premises. The company is a global manufacturer of plastic packaging from beverages to cosmetics - with around 180 production sites worldwide. It is obvious that it is hardly economically feasible to provide trained personnel for maintenance and troubleshooting on site at each production location. Accordingly, a solution was sought that would be able to record and analyze all available production data decentrally in real time and generate appropriate measures in the event of an incident.

The Alpla solution is based on CrateDB, a highly scalable and flexible database, the IoT Data Platform based on it and a central mission control center where all data streams converge and are monitored 24/7. Here, the entire process is displayed transparently at all locations and analyzed according to the stored rules. If the need to intervene in production is detected, the platform automatically generates corresponding notifications to the production staff, together with instructions on how to solve the problem.

Data from a wide variety of sources is integrated into this process - from raw material warehouses, material dosing and production through to visual quality control and the overall status of the process line. Instead of having to maintain a product specialist at each site, the expert in the Mission Control Center monitors the process at all factories and receives automatic alerts from the system. The personnel on site are automatically informed accordingly and provided with information on troubleshooting.

The example shows that IT/OT convergence is currently enabling completely new models of corporate management in production in order to reduce urgent problems such as cost reduction, lack of local skilled workers or waste.

Jodok Schäffler, Head of IoT Data Platform at Crate.io / ag

  • Xing Icon
  • LinkedIn Icon
Advertisement
Advertisement

You might also be interested in

Advertisement
Advertisement
Advertisement
Advertisement
Advertisement
Advertisement
Advertisement

IIoT networking

How production can benefit from AI

Together with AI technology, IIoT networking makes it possible to better control machine parameters and optimize quality with predictive quality. Downtimes and set-up times can also be further minimized. Cloud platforms also make these technologies...

read more...
Subscribe to our newsletter
Advertisement
Back to home