Agentic AI – in other words, AI systems capable of planning, making decisions and carrying out tasks independently – is currently emerging as one of the most significant trends in the industrial application of artificial intelligence. Whilst many AI applications have so far been designed primarily for analysis and support, Agentic AI promises to take things a step further: systems that not only provide information, but also actively control and optimise processes. As a result, expectations within the industry are high. Companies hope this will lead above all to greater efficiency, faster decision-making processes and improved adaptability of complex systems to dynamic environments. Particularly in areas with many interconnected processes, such as production and logistics, Agentic AI can help make decisions in real time, utilise resources more efficiently and continuously improve processes.

Agentic AI is also playing an increasingly important role at KION. On the one hand, we use AI-based agent systems to make internal processes more efficient. This includes, for example, intelligent support for administrative workflows, such as the automated processing of documents, the analysis of operational data, or the optimisation of planning and reporting processes. New opportunities are also arising in production: AI agents can, for example, analyse manufacturing data, identify maintenance requirements at an early stage, or dynamically adapt production processes to changing conditions.

On the other hand, we develop solutions that enable our customers to optimise their own intralogistics processes. Modern warehouse and production environments generate vast amounts of data – for example, from vehicle fleets, warehouse management systems or sensors within the infrastructure. Agentic AI can use this data to autonomously coordinate material flows, intelligently prioritise transport orders or optimise energy and fleet management. One example is systems that independently decide which vehicle takes on which transport order in order to minimise routes, waiting times and energy consumption. In our initial projects, we are already working on integrating such agent-based approaches into our solutions and gradually putting them into practice.

The potential of Agentic AI is particularly significant in the field of intralogistics. Warehouse and production logistics are highly dynamic systems in which many different elements – people, vehicles, IT systems and infrastructure – interact with one another. Agentic AI can help to better manage this complexity and optimise processes holistically. Studies by McKinsey and the World Economic Forum show that AI-supported optimisation in logistics can enable productivity gains of around 10 to 30 percent, for example through better utilisation of vehicle fleets, more efficient route planning or dynamic adaptation to fluctuations in demand. Factors such as energy efficiency, lead times and equipment availability can also be improved.

At the same time, the use of Agentic AI also presents challenges. In industrial logistics, process reliability and dependability are of the utmost importance. AI systems must therefore not only be powerful, but also transparent, robust and traceable. Decisions made by autonomous systems must remain verifiable at all times, particularly when they influence physical processes and machinery. Added to this are issues regarding integration into existing IT and automation landscapes, as well as the quality and availability of the necessary data. Last but not least, the successful deployment of such technologies also requires organisational changes and new skills within companies.

Agentic AI has the potential to mark the next stage in the development of digital intralogistics. The key will be to reconcile the technological possibilities with the high demands of industrial applications – and to develop solutions that are not only innovative but also reliable and practical.

April 2026