AI has long been more than just a trend in logistics. The BME Logistics Study 2025 shows where companies are already benefiting today, which technologies quickly pay off and which investments directly increase transparency and efficiency.

Digitalization is becoming the focus – AI is becoming an accelerator

Planning, production and logistics are hardly conceivable without digital technologies. Since the end of 2022, AI has been gaining pace – driven by technological leaps and regulatory impulses such as the EU AI Act. The BME Logistics Study 2025 provides a practical overview: Which technologies are known, which are in use and where do measurable effects arise?

What was investigated?

236 specialists and managers from various industries and functions (with a focus on SCM and logistics) were surveyed about 15 digitalization technologies and AI use cases along the supply chain operations reference processes (SCOR). The results make it clear where companies are already working in normal operations – and where there is potential.

Technologies at a glance

The study shows how extensively central digitalization technologies are already being used today – and where companies are still hesitant:

• High awareness: AI, robotics/automation, autonomous vehicles; Digital twins, IoT, cloud/API and VR/AR are also widely present.
• In productive use: Cloud/API and robotics/automation (around 60 percent each).
• Catch-up: AI, big data, advanced analytics – about half are testing or planning.
• Niche status: Edge computing, bionic enhancements, quantum computing.
• Next Gen Wireless (5G, Wi-Fi 6/7, LPWAN): important enabler for transparency and IoT.

Where benefit arises

Digital technologies primarily contribute to classic SCM goals:

• Cost reduction, time saving, quality improvement are the top effects.
• Robotics/Automation scores particularly well when it comes to costs.
• 3D printing improves costs, time and flexibility.
• Big Data, Digital Twins and IoT support risk identification and responsiveness.

Drivers and brake pads

• Driver: clear business benefits, increasing competitive pressure, customer expectations, open culture and committed employees.
• Obstacles: lack of skilled workers (top brake), high investments, data quality and mindset/know-how.

The look ahead

The next two years will be crucial in determining how quickly AI scales in logistics:

• AI: Over 80 percent plan to use it within two years. Advanced analytics follows closely behind.

• Quantum computing: no short-term plans; relevant in the medium term.
• Blockchain: low priority; only every fifth company plans for the short term.

AI in practice – use cases along the SCOR processes

The SCOR processes show where AI is already having an impact today – and where potential still remains unused:

• Plan: High potential for demand forecasts, simulations and inventory optimization; Productive use is still low, tests are increasing.
• Source: Automated orders and early risk detection are promising, but have not yet been implemented much. Contract analysis using NLP makes sense, supplier scoring is under development.
• Make: Predictive maintenance, production planning and visual quality control are the furthest. Energy consumption optimization follows.
• Deliver: Route optimization is at the forefront – more than one in five companies are already using AI; ETA forecasts and chatbots to follow.
• Return: Lower priority, but opportunities with automatic sorting and return forecasting.
• Enable: Automated reporting/KPIs has the highest potential; Master data management with AI/NLP remains a key enabler.

Five recommendations for logistics professionals

From the results, clear action steps can be derived that pragmatically advance digitalization and AI:

1. Understanding technologies: Know trends, benefits and limitations – beyond GenAI.
2. Clarify roles: Define responsibilities, resources and partner structures.
3. Prioritize specifically: Evaluate business case, process maturity, database and ROI for each use case.
4. Think AI broadly: Combining ML, computer vision, NLP and agents sensibly.
5. Strengthen master data: Clean data is mandatory – AI can provide support, but it does not replace basic work.

What does that mean for everyday life?

Start where benefits and feasibility come together: automated KPIs, route optimization, predictive maintenance, inventory optimization or automated orders deliver quick effects. At the same time, it is important to strengthen data quality, skills and change management. Machine learning is currently the most pragmatic entry point – other AI sub-areas specifically complement it. In this way, AI is gradually becoming an integral part of resilient value creation.