Artificial intelligence (AI) is revolutionizing numerous industries. In intralogistics, the main focus is on optimizing processes, expanding automation and developing innovative solutions. The trend here is increasingly moving from so-called weak to strong AI.
2025-04-17
In large warehouses, every second is valuable: Speed and efficiency mean lower costs, shorter delivery times and satisfied customers. The KION Group is therefore systematically using artificial intelligence (AI) to further accelerate processes and make even better use of resources.
The decisive factor here is the difference between "weak" and "strong" AI. What's it all about? "We speak of strong or true AI when a system is able to learn independently, understand complex relationships on its own and find solutions to certain challenges without human intervention," says Christoph Hock, Head of Sales Software Solutions ITS EMEA.
According to Hock, strong AI is becoming increasingly important in intralogistics in order to remain competitive in the long term and drive innovation.
Weak AI: Systems for specific, clearly defined tasks without actual understanding. Examples: Chatbots, speech recognition, recommendation systems. Features: Rule-based, limited adaptability, no independent further development.
Strong AI: Systems that can learn, understand, abstract and react flexibly to completely new situations – similarly to humans. Strictly speaking, there is currently no fully developed strong AI. However, technological precursors such as generative AI, genetic algorithms and autonomous robots are showing the first signs of this: They are partly self-learning, can recognize patterns, make decisions and adapt to changing conditions. Properties (in the target concept): Self-learning, generalizing, capable of making decisions, highly adaptable.
"Generative AI (GenAI)" is an important development step in the direction of strong AI.
"Generative AI (GenAI)" is an important development step in the direction of strong AI. Even if, strictly speaking, it is not strong AI, it already exhibits several characteristics of strong AI, particularly in the area of adaptability and pattern recognition – as can be seen in the example of our warehouse management systems (WMS).
These must always be customized, as every warehouse is unique. Generative AI can significantly accelerate this adaptation process. "With GenAI, we can do in a few seconds what would take human developers several days," says Dr. Johannes Hinckeldeyn, Director of Advanced Core Technologies at KION.
Specifically, artificial intelligence can automatically configure a customized WMS based on a detailed description of all warehouse processes and structures. This saves the KION Group time in three ways: in the planning phase, during implementation and during operation. In addition, AI also handles test runs and documentation automatically.
The standardization of the adaptation process guarantees high quality, reduces sources of human error and makes it possible to react more quickly to changing requirements. KION thus gives its customers a tangible competitive advantage: Every little bit of time saved in the warehouse means higher availability and throughput.
AI makes it possible to control a highly complex warehouse environment intelligently in real time
AI methods are now also being used in warehouse operations, for example genetic algorithms that optimize the allocation of transport orders in fractions of a second.
Instead of processing orders rigidly one after the other or spending time trying out every possible allocation option, the algorithm finds a solution that is among the best two to three percent of all possibilities in just a few seconds. The results are impressive: "We have shown in simulations that the genetic algorithm achieves time savings of up to 20 % compared to other algorithms, depending on the warehouse conditions," explains Hinckeldeyn.
In concrete terms, this means that seconds or even minutes are saved per vehicle movement, which add up to whole days over the course of a year. This micro-optimization of each individual order adds up to considerable efficiency gains in the overall process – all thanks to (almost) strong AI.
The combination of GenAI and genetic algorithms not only opens up new possibilities for designing logistics processes – existing processes can also be improved. This makes it possible to control a highly complex warehouse environment intelligently in real time, resulting in higher capacity utilization, greater throughput and ultimately more satisfied customers.
These belong to the family of evolutionary algorithms – AI processes that are inspired by the principle of biological evolution. Instead of programming a solution in advance, the AI develops the solution itself by generating many possible solutions and subjecting them to a "survival of the fittest" as in nature.
Artificial intelligence opens up completely new possibilities for driving innovation in intralogistics.
At the KION Group, we are already using artificial intelligence in a large number of projects. Examples of this are:
• Energy management for electric forklifts: In cooperation with software developer ifesca, KION has developed an . The AI predicts the specific energy requirements of each individual forklift truck and helps to plan optimum charging times, avoid peak loads in the power grid and thus significantly reduce energy costs. At the same time, the more efficient use of the charging infrastructure can support customers' sustainability goals.
• Digital twins for warehouse optimization: KION is working with tech company NVIDIA and consulting firm Accenture to . The joint project was presented for the first time at the CES technology trade fair (USA). The focus is on the use of the NVIDIA Omniverse to create digital twins in which AI models can simulate a wide variety of scenarios – for example for route optimization, resource utilization and process planning. A first application case was demonstrated at LogiMAT (Germany) in which autonomous vehicles interact with each other in a virtual environment – an intermediate step on the way to digital mapping of complete warehouse processes. A simulation of the Tech Center in Grand Rapids – a development site that serves as a realistic test field for AI-supported optimization – was also presented at the KION Group Technology Conference.
Artificial intelligence opens up completely new possibilities for driving innovation in intralogistics. At the KION Group, we are pioneers in this field. Our customers and partners not only benefit from already established AI solutions such as optimized processes and lower costs, but can also rest assured that KION is actively shaping the technological future of intralogistics – second by second.
