Artificial intelligence (AI) is experiencing a noticeable boost, but what is the status of AI in logistics? 15 percent of companies in the German economy now use AI; a year ago it was only nine percent. At the same time, the proportion of those for whom the use of AI in their own company is not an issue has fallen significantly: from 64 to 52 percent. A good two thirds (68 percent) consider AI to be the most important technology of the future. These are the results of a study commissioned by the digital association Bitkom from 2023.
The topic of AI is also slowly gaining momentum in logistics. Professor Andreas Rükgauer, who researches the application of AI in value creation at the Würzburg-Schweinfurt University of Applied Sciences (THWS), is currently primarily mentioning “the low-hanging fruit” that can be harvested in logistics with the help of AI applications: primary fields of application For beginners, the automated reading of receipts and document management, or even the optimization of loading density, are important. Image recognition with neural networks also works very well. “In contrast, AI-supported route optimization is very complex and more for advanced users,” says Rükgauer.
What role will AI play in logistics?
There are numerous other areas of application in which AI can be used in logistics. Although AI solutions are not yet “standard products” and must be adapted to individual circumstances, the IT company Arvato Systems, which is part of the Bertelsmann Group, identifies the following scenarios:
- Error detection in master data: Master data in the sense of logistics is data that contains operationally relevant information – for example about products, orders, suppliers and customers. Errors in this data lead to reduced efficiency or even process interruptions. An AI can detect anomalies in this data, which can then be corrected.
- Path-optimized article positioning: One of the biggest challenges in the warehouse is keeping travel times as short as possible. This can be achieved by optimizing item positioning in the warehouse as well as path-optimized picking. AI helps to optimize the picking of items in time.
- Personnel deployment planning: Demand-based planning of personnel deployment is crucial for smooth yet cost-efficient operational processes. Based on a wide variety of parameters such as historical process information, order and order history, and information on process times, AI can help to achieve holistic cost optimization.
- Inventory check: Too little stock can lead to delivery bottlenecks, while too much stock can lead to avoidable storage costs. AI in logistics can optimize inventory checking: based on historical order data, the AI predicts the expected orders and thus creates the basis for optimal inventory planning.
- Replenishment control: There must always be enough goods in the picking zone; replenishments that are carried out too often hinder the processes and result in unnecessary work. Based on incoming orders, AI makes it possible to predict which quantities of goods must be available in the picking zone at a certain time.
- Position detection in warehouse architecture plans: For route-optimized item positioning, a digital location plan with coordinates of the individual storage locations is required. Creating such plans manually is time-consuming and often takes several days. AI can also provide valuable help here and minimize the time required.
- Chatbots: Whether in the warehouse or on the ramp – information must be constantly retrieved in logistics processes. The chatbots known from the e-commerce environment also provide good service here. AI-supported, they help employees to quickly and securely receive the relevant information via a mobile device.
- Container management: Container management is another field of AI in logistics. Optimal planning and utilization of the container arsenal ensures that enough suitable containers are always available at the right time and in the right place to process the orders. At the same time, the container inventory is optimized to avoid overhangs.
- Dock & Yard Management: Yard logistics can be extremely costly if not managed optimally – dock & yard management is crucial to a frictionless logistics operation. Combined with image recognition for identifying trucks, trailers and containers as well as data on the current volume of deliveries and deliveries and the schedule, an AI determines the best possible placement of the trucks.
AI in logistics is already being used by large companies
AI applications have already been implemented or are being tested in some large companies. The forwarding and logistics company Duvenbeck for example, is participating in a research project on the use of AI in logistics and production. Duvenbeck supplies extensive, detailed data from the company’s own transport management system (TMS) via various interfaces to two project partners from the manufacturing industry. On the one hand there is the agricultural machinery manufacturer Claas and on the other hand there is the trailer manufacturer Schmitz Cargobull, both long-standing partners of Duvenbeck. With the data prepared by Claas and Schmitz Cargobull and analyzed using AI, Duvenbeck, as a logistics partner, can make more precise predictions about shipment volumes, vehicle and storage space utilization, personnel deployment and many other steps in the supply chain.
Logistics service provider Badger uses the possibilities of AI in general cargo logistics. The “@ILO” technology, developed together with Fraunhofer IML and implemented in two Dachser pilot branches, is based on the idea of the digital twin and fully automatically creates a current, digital image of all packages, assets and processes in the transshipment warehouse. AI-based algorithms interpret the data recorded every second by hundreds of cameras on the hall ceiling. The result is impressive: individual process flows between incoming goods and outgoing goods accelerated by a range of 15 to 35 percent. For example, there is no need to manually scan barcodes or take daily inventory of packages.
The railway subsidiary DB Schenker uses intelligent algorithms, among other things, for the optimized allocation of areas to the terminals, for personnel and resource planning, for optimized loading of trucks and containers and for dispatch planning. In addition, the company has introduced computer vision technology including automatic license plate recognition in yard management.
Does AI in logistics cost jobs?
Directly linked to the rise of AI is of course the question of what it means for people’s work. Definitely a lot of change. For example, ChatGPT inventor Sam Altman himself admitted in an interview that AI “could cost a lot of jobs” in the coming years. At the same time, new ones are likely to emerge.
The impact of AI on logistical job profiles is also unlikely to be ignored in the long term. “Every job profile will undergo one or more changes in the next, let’s say 20 years, in some cases even so far-reaching that the job profile no longer exists,” says Christian Kille, Professor of Trade Logistics and Operations Management at the Würzburg University of Technology -Schweinfurt (THWS). In principle, everyone would have to be prepared to continue their training and adapt to new conditions. “I don’t know how the dispatcher will do his job in ten years – but it will definitely be different than it is today,” says Kille.
