Digital assistants in logistics practice
By Andre Kranke |
7 minutes read
30/05/2025
Artificial intelligence (AI) is becoming an everyday technology. It opens up considerable potential for logistics, too—and that goes beyond the analysis of large volumes of data. As a practical assistant, AI supports employees in decision-making or takes over monotonous routine tasks. Andre Kranke, Head of Corporate Research & Development at DACHSER, explains how AI is already being used in groupage logistics and what potential it has.
Is artificial intelligence (AI) a technology of the future? The answer is both yes and no. Yes, because we’re a long way from exhausting AI’s potential, and no, because AI applications have long since become a part of our everyday lives—through facial recognition on smartphones, chatbots, or translation services on the internet. AI is also used in logistics more often than you might think. It powers applications in areas such as the prediction of shipment quantities, the control of material flows, and the support of administrative processes.
It’s also a fact that AI isn’t yet fully developed in many areas. This simulation of human intelligence is nothing more than complex math with a lot of probability calculations. That’s why, depending on the data quality, it inevitably produces errors—and these need to be minimized, especially in logistics for sectors with its high demands on safety and quality. Error reduction needs to happen both in so-called AI agents, which act independently in non-critical secondary processes, and in AI assistants, in which humans always control and monitor. And that explains why from today’s perspective, there can be no logistics without people. The final decision must always be made by a human, especially when business risks are involved and there’s no scope for tolerating a certain error rate.
But all the same, logistics companies that aren’t yet looking intensively into AI will be eliminated from the market in the medium to long term. After all, customer requirements and the complexity of the conditions aren’t getting any less demanding—and that’s before we factor in the growing lack of qualified personnel. DACHSER is already using AI applications in various areas, including warehouses, transit terminals, and offices, to provide its employees with the best possible support in decision-making, increase efficiency, and minimize bottlenecks. This helps compensate for the lack of qualified personnel and guarantee a high level of quality in the long term.
New, unexpected possibilities
More than six years ago, in the DACHSER Enterprise Lab—a research and development laboratory at Fraunhofer IML in Dortmund—DACHSER began developing algorithms that can, for example, forecast inbound volumes at branches up to 25 weeks in advance in order to support seasonal capacity planning. After all, the crucial factor for achieving efficiency and quality in logistics is the ability to plan. Here, too, AI can really make a difference, as demonstrated by DACHSER’s first machine learning project: PAnDA One. This is an acronym of Predictive (P) Analytics (An) DACHSER (DA), where “One” denotes that it is the company’s first machine learning project.
AI algorithms are also used in the @ILO digital twin, which identifies, localizes, and measures packages in groupage warehouses in real time. It uses cameras on the ceilings to create an exact digital map of all movements and procedures within the warehouse. This increases transparency and makes for a better overview. At the same time, it does away with manual processes such as scanning the goods, making certain unloading processes up to 30 percent more efficient. DACHSER will gradually roll out the @ILO digital twin across Europe in the coming years, with at least six new locations planned for 2025.
In the @ILO terminal, packages are identified, localized, measured and recorded in the transport management system fully automatically as they enter, stay and leave. Manual scanning of barcodes and additional labeling of the packages is no longer necessary.
Autonomous AGV robots in the warehouse
Transit terminals aren’t the only places where using AI makes sense. Take automation in the warehouse: self-driving transporters called automated guided vehicles (AGVs) are now in use in eight DACHSER warehouses in Germany. Also known as autonomous mobile robots (AMRs), they map their surroundings with sensor systems like cameras, lidar, and radar and find their way around with the help of AI. This lets them autonomously perform simple repetitive tasks. The driverless transport systems make their own way through the warehouse, for example to store and retrieve pallets at ground level. They communicate with each other and swap driving orders if another vehicle is in a position to reach the destination faster. If there’s an obstacle in the way, the vehicle brakes. This way, the robots almost never travel unnecessary distances, meaning they work very efficiently. At the same time, safety sensors ensure that the vehicles don’t get into any accidents. It’s fascinating how well this works.
Precise navigation, swarm intelligence, and seamless integration into the IT systems deliver a major increase in the efficiency of day-to-day logistics—and employees are also pleased to have a “robot colleague” to lighten their workload. DACHSER uses the AGVs in mixed operation. That means they share the driveways with human-driven vehicles. Complete automation makes little sense, as it would reduce flexibility. It’s now a case of finding the optimum combination of humans and machines to add the most value..
Researching togehter
What can we expect to see in the future? In the robotics industry, experts are experimenting with foundation models as a means of communicating with and controlling autonomous vehicles. This would enhance robots’ ability to perform complex tasks such as natural language processing, image and object recognition, and autonomous navigation. These models also allow robots to learn from vast amounts of data and adapt to new environments and tasks, which in turn means they offer greater flexibility and a wider range of applications. And it won’t be long before we see if the autonomous vehicles used in warehouses can be controlled more intuitively and efficiently. Intensive research is being carried out worldwide.
DACHSER is also increasing its clout in AI research. In the first quarter of this year, the logistics provider expanded its research partnership to include the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS in Sankt Augustin. Fraunhofer IAIS is a leading scientific institute in the fields of artificial intelligence (AI), machine learning, and big data in Germany and Europe. Its almost 400 employees help companies optimize products, services, and processes, and they offer support in the development of new digital business models. Having this new partner in the DACHSER Enterprise Lab lets us further strengthen our expertise in the field of artificial intelligence.
Further research will yield a whole new range of potential applications for AI, which must be trained with specific internal company data, especially for special logistics processes and solutions. At the same time, consideration must be given to the costs, especially of AI models that require considerable computing power, as well as to compliance with the EU’s new legal framework for AI applications as laid out in the AI Act.
To summarize, AI offers logistics many opportunities to do things that were simply not possible before. However, it’s not really “intelligent,” but rather a tool based on higher mathematics, large amounts of data, and computing power. And it isn’t the optimal solution for all digital problems; conventional programming is often still the better way. The task now is to find the right mix between the use of standardized AI applications and in-house developments, and then to adapt these to our own requirements.
