AI, logistics, and what comes next
Thoughts on computer vision, scalable AI, and the future of forwarding, distribution, and warehousing — from the people building it.
The gap between theory and practice in supply chain optimisation
Even the most sophisticated supply chain optimisation models are only as good as the data behind them. A 10% data inaccuracy can render results meaningless — making real-time visibility a prerequisite, not an afterthought.
Episode V – Return of the Data
With the missing conversion ratio identified and automated data capture in place, a Nordic furniture retailer eliminated overbooking, reduced wasted transport capacity, and realised €1.3 million in annual savings — all without heroics or guesswork.
True value unlocks through process understanding
The real unlocking of value comes from a deep, intimate understanding of the working process — not from technology alone. A reflection on Episode V of Ivan Markov's supply chain visibility series and what it means to truly understand how transport booking works.
The logistics industry is at a fascinating crossroads
Four practical AI applications delivering measurable results today: real-time visibility over prediction models, dimensional accuracy reducing shipping costs by 18%, fast-but-simple over abandoned-complex, and human expertise amplifying AI effectiveness.
Episode IV – The Missing Piece
After identifying where money was leaking, the team stopped analysing and started listening — discovering that a static, never-updated Conversion Ratio was the missing link between insight and €1.3M in realised savings.
Container optimisation: beyond 'fill and ship'
'Fill and ship' is no longer enough. Real-time visibility into what's actually in each container — versus what the manifest says — helps logistics leaders improve utilisation by 15–25% and reduce costly surprises at destination.
Sometimes the biggest wins come from solving the smallest inefficiencies
Seven minutes of manual admin per container doesn't sound like much — until you multiply it across hundreds of daily operations. A global logistics company achieved 6% warehouse productivity gains and increased shipping capacity without adding headcount.
Episode II – The Proof of Concept Strikes Back
The Inspector solution captured real images of truck loads at a Nordic furniture retailer, revealing systematic overbooking and underutilised capacity. Just 200 data points pointed to €700K–€1.4M in potential annual savings.
Episode I – A New Approach
Every logistics transformation starts with challenging a belief. Approaching a Nordic furniture retailer not with a pitch deck but with curiosity, the team aligned on a single shared mission: only ground truth from the warehouse floor.
Route planning: the art and science of modern logistics
Route planning has evolved far beyond A-to-B navigation. Winning logistics companies blend algorithmic optimisation with real-time visibility and human expertise — achieving better service at lower costs across their entire supply chain.
Sentispec and DHL: Unlocking AI-powered container intelligence
Sentispec and DHL have partnered to deploy an AI-powered solution for a global heavy equipment manufacturer — unlocking transparency, identifying hidden inefficiencies, and driving smarter logistics decisions through real-time container tracking.
Smarter loading equals smarter logistics
At DHL, we believe smarter loading equals smarter logistics. With Sentispec's load analysis, we're boosting container fill quality, cutting down on empty space and damage risk, and saving real money for our customers while delivering the same high level of service.
Cutting CO₂ through AI-driven trailer utilisation
At DHL, cutting CO₂ isn't just a goal — it's a strategy. Our collaboration with Sentispec enables real-time monitoring to ensure trailers operate at full capacity, reducing wasted trips and environmental impact while maintaining service quality.
How to scale predictable AI
The use of AI in logistics can be cumbersome to scale because of the need for extensive training and huge data loads. By prioritising only needed information and ignoring the rest, we can reduce a 10 MB problem into 0.5–1 MB — vastly reducing cost and increasing scalability. The key to success is determining what information is really needed, and ignoring the rest.
The disappointment of autonomous AI
Autonomous AI systems that promise full independence often disappoint in real-world logistics deployments. Understanding why — and how to design AI that delivers predictable, scalable results instead — is the difference between a pilot and a production solution.
Musings on human vs. computer vision
Human eyes excel not because of mechanical superiority, but because of the brain behind them. By strategically positioning cameras and processing visual data in real time, organisations can transform continuous observation into actionable business intelligence — at a fraction of the cost of manual monitoring.
End-to-end supply chain visibility
End-to-end supply chain visibility is the ultimate goal in logistics — but RFID and scanning portals are too expensive to deploy everywhere. Computer vision offers a cost-effective alternative: leveraging cameras already on forklifts, gates, and smartphones to track goods from source to destination in real time.