Artificial Intelligence (AI) is playing an increasing role in the digitisation of the logistics sector. During ICT&Logistics and Logistica Next, Alexandros Visser participated in the sector panel "Digitisation of internal logistics: grip or dependence?" Martijn Graat of LogistiekMet... interviewed him afterwards for the LogiNext Talks series. In this article, you will find the key insights and practical lessons from the interview.
1. AI won't work without people
A common mistake is that companies want to fully automate AI. The interview revealed that many AI projects get stuck in the pilot phase.
Why.
AI is not traditional automation. It is less predictable and needs human supervision to function properly.
Takeaway:
Involve humans in AI workflows for control and adjustment.
2. AI has no human understanding, but it can recognise context and respond to it consistently.
AI has no human understanding, but it can recognise context and respond to it consistently.
Practical example:
If you start a sentence with "When the cat is away from home...", AI automatically adds: "...the mice dance on the table".
Takeaway:
Use AI as support, not as a replacement. Always check the output.
3. AI applications in logistics
According to Visser, the biggest opportunities lie at the front end of logistics organisations:
- Automating quotation and order requests
- Accelerating customer communication
- Dynamic planning for unexpected changes
- Supporting planning and decision making through AI agents and smart systems
Takeaway:
Start AI where flexibility, speed and lots of unstructured data are involved.
4. Data remains yours
A key concern: what happens to business data?
Visser advises:
- AI hosting in your own cloud environment (e.g. via Microsoft).
- On-premise solutions are possible, but in practice many organisations opt for (private) cloud solutions because of scalability and management.
- This is how you keep more control over your data, provided governance, access management and security are properly set up and you comply with AVG guidelines.
Takeaway:
Keep control of your data when using AI.
5. Start with processes, not technology
Many companies start AI projects from a technical angle. Visser advises: start with the business processes.
Roadmap:
- Map processes
- Find bottlenecks
- Determine where AI adds value
Takeaway:
Think from your business, not hype. Traditional automation is often cheaper and more effective for predictable tasks.
When do you use AI (and when not)?
AI is powerful, but not always necessary:
Use AI as:
- Making processes variable
- Language plays a big role
- Flexibility is important
Don't use AI if:
- Making processes predictable
- Simple automation is enough
- An AI solution is more expensive and complex than traditional automation
Conclusion: AI is a tool, not a substitute
AI is all about cooperation between humans and machines. Companies that find this balance:
- Combining human knowledge with AI
- Maintain control over data, integrations and processes
- Make smart choices about where AI adds value
Want to see more interviews with speakers from ICT&Logistics and Logistica Next? View all interviews from LogiNext Talks here.
Want to be kept up to date with the latest logistics news? Then subscribe to LogiNews.
