comPETence magazine | ONE:22 | Inspection and Digitalization
Linatronic AI - Deep learning in inspection technology
Anyone who works with empty-bottle inspectors knows that not every bottle that the inspector rejects actually has a defect. In some cases, it might simply be water droplets or a bit of foam still clinging to the bottle after cleaning. Since conventional systems can’t always distinguish these from contaminants or damage with 100 percent certainty, they tend to err on the side of caution and reject the container. As a result, in every production shift, countless perfectly usable bottles land in the trash.
To change that, Krones has taken the evolution of its inspection technology to the next level. The new Linatronic AI employs deep learning software to automatically detect and classify anomalies, making it much smarter and more efficient than its conventional peers.
The next rung on the evolutionary ladder: artificial neural networks
Deep learning is a technology that enables machines to do what we humans do naturally: learn from example. But there is one big difference: a machine can use this ability many times more efficiently than humans can.
The foundation for deep learning is an artificial neural network (ANN). The ANN can be described as a complex system of multiple consecutive filters. The images captured during the inspection process are fed through these filter layers, one after the other. Each layer extracts a different characteristic of the image. Since one filter’s output becomes the input for the subsequent filter, the complexity of an image’s characteristics can be increased almost infinitely. The chain ranges from simply identifying darker or lighter pixels all the way to classifying very specific objects such as beads of water.
Trained using thousands of images
To ensure that the Linatronic AI applies these filters with the necessary precision in practice, it is trained ahead of time with pre-classified example images. In this way, its neural network learns to filter out and interpret the relevant image characteristics. The same is true for machines that is true for us humans: the more intensively you train, the better the results. Therefore, Linatronic AI’s neural network was continually fine-tuned using thousands of example images until it could accurately distinguish water droplets from other anomalies – with a reliability rate of over 99.9 percent.
As a result, waste due to false rejects is no longer an issue. The time-consuming process of configuring the inspector during commissioning is also a thing of the past. The neural network doesn’t require manual calibration to local conditions. Instead, the Linatronic AI is delivered fully trained and ready to start work.
Krones continues to pool its digitalization activities
Digital issues have been assuming an increasingly pivotal role in the Krones Group. It is a trend that, for some time, has been affecting not only internal group structures but also the product portfolio for Krones customers, who since 2021 have benefited from overarching service level agreements and the central Krones.world platform that pools all Krones services offered for beverage producers. In 2021, Krones was ranked among the champions of digital transformation for this holistically conceived change process which is now being rigorously pursued.
The enterprise is continuing on this path and has set up a new organisational unit – Krones.digital – that in January 2022 brought together all groupwide resources in the fields of digitalization and automation in order to adopt an overarching approach to developing solutions primarily for gathering, analysing and utilising production data.
Global human power for a digital future
What is special about it is that Krones.digital is not a department in the usual sense of the word but a community consisting at present of roughly 450 experts, all of whom work for the Krones Group yet come from a wide range of specialisms, from different companies and different continents. The unit thus forms a matrix organisation within the group structure. The matrix design ensures open communication between all the experts involved, transcending global specialisms and company borders, and facilitates brainstorming and cooperation along the entire value chain, from product development and project management right through to supporting customers in their day-to-day business. The experts all across the globe use the same technologies for this and are establishing a shared tech stack. That is how Krones devises the digitalization process for production operations across four levels – starting with the machine, then line control, a central Cloud platform with digital services, right through to the SAP platform where the Krones Group offers further applications designed to control digital beverage plants.
To sum up: Krones’ new unit deploys a flexible, fast-response crew with a broad range of specialisms, whose members work together in pursuit of a single mission: to lead the food and beverage industries into a digital future. The community sees its role as designing an efficient, sustainable beverage production operation “managed by Krones” that will enable customers to focus on their own business activities.
Krones.digital – a growing community
This type of cooperation in a cross-specialism organisational unit is new to Krones. But having established and consolidated it over recent months, Krones is now moving on to the next step and is looking to hire new employees with digital expertise. “We set great store by a pleasant working environment and a certain team spirit – with open, straightforward cooperation, direct communication channels and streamlined decision-making being our top priorities. We have fostered this team culture over recent weeks and months and are now looking forward to seeing Krones.digital grow with its new faces,” explains Severin Diepold, Head of the Corporate Digitalization and Automation subsection within Krones.digital.
The comPETence center provides your organisation with a dynamic, cost effective way to promote your products and services.
Find our premium articles, interviews, reports and more
in 3 issues in 2023.