AI in Logistic Services: Not a Hype, but a Necessity

By Jan van Elderen, Logistics & IT Lecturer/ BUas AI Pioneer
 

Not because you have to, but because it pays off

The logistics sector is under structural pressure and faces a great deal of uncertainty. Think of staff shortages, disruptions in international logistic chains, the current tariff war, stricter sustainability requirements, and customers who have increasingly high expectations of delivery times and service. AI is not a panacea, but it can help to deal with all these challenges in a smarter way.

Take route planning. AI models can combine real-time traffic data, weather and history to automatically create feasible schedules, more effectively than manual planning. Depending on where the priority lies, this results in shorter routes, lower fuel consumption or higher delivery reliability. It is no science fiction, but it is already possible today. Or think of the predictive maintenance of a vehicle fleet. By combining sensors with AI algorithms that recognise patterns in malfunctions, you know when a vehicle is at risk, and you can take this into account in your planning. You prevent expensive (corrective) downtime and extend the service life of your equipment. Furthermore, maintenance can be planned more effectively in advance.

There are also opportunities in the warehouse. AI-driven systems can optimise order picking, manage stocks more intelligently, better predict what needs to be purchased and when, and support employees with visual recognition during quality checks. These types of applications are particularly valuable in places where repetitive work is becoming increasingly difficult to fill. With recent developments, such as AI-driven WMS systems in which employees can ask questions to the system via a chat function, technology is also becoming more accessible and interactive than ever before, enabling companies to onboard new employees, often as temporary workforce, in a shorter period of time.

Look beyond only the ‘hard’ logistic processes

AI applications within logistics are mostly gaining ground at the intersection of customer contact and commercial efficiency. Think of automatic email handling or chatbots that handle frequently asked questions or order status requests – quickly, accurately and available 24/7. Such tools reduce the workload and increase response times, provided they are properly set up. AI also offers opportunities in the commercial sphere, for example with automatically generated price proposals based on previous quotations and current costs. And AI driven feedback analysis reveals trends in customer reviews more quickly, which helps to improve processes in a targeted manner. 

However, not everything can be automated. Human contact remains essential in unusual situations or when dealing with more complex customer queries, and this makes it even more valuable. By automating standard questions in a smart way, you free up time for more personal interaction where it really matters.

Why AI initiatives fail: lessons from practice

Unfortunately, all that glitters is not gold. AI initiatives regularly fail—and often for recognisable reasons. Major pitfalls? Poor data, or no access to the right data. Without good, structured and reliable data, AI remains dumb. The algorithm is only as smart as the information you put into it. And, when working with data, you must remain compliant of course. Data security is therefore an important part of your initiative.

In addition, the possibilities of AI are often overestimated. Companies expect immediate profits or spectacular breakthroughs, whereas in reality it is often a matter of gradual improvements. Think in terms of months rather than weeks, and realise that experimentation, learning and adjustment are part of the process. AI requires a different way of working and thinking. Without support from employees, or without people in your organisation who understand what the technology can (and cannot) do, it will remain a toy for the IT department. But that's not enough. AI requires collaboration between IT and ‘the business’/operations. In conversations with various logistic service providers, one thing is always emphasised: AI only really works when it is embedded in the broader context of your organisation, including change management and process intelligence.

What will be the first step? Just begin!

The first step does not have to be big. In fact, it should be small. Do not wait for the perfect solution or the ultimate plan. As Pieter Zwart of Coolblue aptly puts it: “Just do it.”  Start with a specific problem in your organisation that you encounter on a daily basis.

Give a small, multidisciplinary team space, experiment, fail and learn. In logistics, in particular, learning by doing works as there is enormous pressure to improve every single day. And perhaps even more importantly: bring in new, fresh minds. People who did not grow up with Excel, but who can work with tools such as PowerBI and are familiar with modern programming languages such as Python. Colleagues who think in terms of prompts rather than emails. Who are curious, not afraid of technology, and who look at your processes with a fresh perspective. They do not need to know everything about logistics—they will learn that. But they bring the mindset needed to truly innovate. 

Companies such as IG&H see in practice that it is precisely these small-scale, targeted projects that quickly lead to tangible results—provided that attention is paid to the human factor and the process. It is important not only to introduce the technology, but also to give ‘ownership’ to the employees who work with the processes on a daily basis.

AI is not optional, but strategically smart   

For logistic service providers, AI is not a luxury or a fad, but a logical step towards future-proof business operations. Start small, learn quickly, fail smartly, and scale up. The technology is there, the benefits are real, and the urgency is greater than ever. And as Daniëlle van Hout puts it: “Without process intelligence, AI does not deliver the expected results.” AI does not work in a vacuum. It works within your processes, with your people, and at the service of your customers. All you need is the courage to get started. 

This does not start with experienced employees. Students, in particular,  demonstrate how quickly you can create concrete value with a fresh perspective and smart tools. At Breda University of Applied Sciences, for example, Logistics Engineering students already work on practical assignments at the intersection of logistics and data technology in their third year. For example, one student developed an AI-driven forecast model that predicts the number of cargo metres required per day with 86% accuracy. Another student linked systems to create a dashboard that provides insight into the profitability per trip and per customer. And yet another built a cross-selling model that optimised both the layout of the warehouse and turnover. These examples show us: learning by doing works. And students can make a real difference with up-to-date knowledge, creativity and digital skills.