The Evolution of Transportation Management: Increased Efficiency by Smart Automation

Martijn Graat
Martijn Graat

Does logistics matter? It’s the opening question of every Does Logistics Matter? podcast episode, and when Elmer Spruijt, VP Transportation Management EMEA from Descartes, settled in to talk about transportation management systems, the answer came with the quiet confidence of someone who has watched this industry transform from the inside.

Not because TMS is suddenly in the spotlight, but because most companies are still only scratching the surface of what these systems can actually do.

From Back Office Calculation to Living System

Ask most people what a TMS does, and they’ll describe a planning tool. Something that figures out how to move goods from A to B as cheaply as possible. That description was accurate twenty years ago. Today, it’s about as complete as describing a smartphone as something you use to make calls.

A modern TMS is connected, dynamic, and constantly active. It pulls data from telematics devices in trucks, pings GPS every few minutes, and connects with ports, customs authorities, rail providers, and ocean vessel tracking systems. It calculates ETAs using traffic, rest times, weather, and historical performance on specific lanes. It exchanges documents automatically and tracks carbon emissions. It doesn’t just produce a plan; it manages its execution in real time and responds when reality diverges from it.

Thirty Years of Invisible Infrastructure

None of that intelligence works without connectivity, and connectivity is harder to build than it looks.

Descartes has been at it for over thirty years and now has more than 200,000 companies connected to its network. Yet they still add 2,000 new carriers every month. That number is a useful reality check for anyone who thinks the logistics ecosystem is nearly fully wired up.

The network matters beyond just visibility. When a shipper switches carriers after an annual tender review, a system without a network requires a brand-new technical integration. A system with a network either already has that carrier on board or adds them quickly. What looks like a lower price per container can quietly come with a hidden connectivity cost that never makes it into the procurement spreadsheet.

When the Truck Knows More Than the Driver Admits

One of the more striking moments in the conversation is when Elmer describes double-tracking, using both a truck’s onboard telematics unit and a driver’s smartphone app simultaneously. In most cases, it’s simply a way to get richer data. But in fraud cases, it becomes something more powerful.

If a truck is stolen or cargo is diverted, a single data source can be manipulated to make it appear the vehicle is still moving normally. If you only have one source, you have no way of knowing. If you have two, the discrepancy becomes immediately visible. A truck showing movement in one system and a stationary signal in another is a flag that something is wrong.

It’s a good example of how data redundancy in logistics isn’t just about backup. It can be the difference between catching a problem and missing it entirely.

The Phone Call That Changes Everything

Getting data into a TMS requires drivers to be connected. That sounds straightforward until you consider the scale involved: hundreds of thousands of drivers across countries and languages, with varying levels of comfort with technology.

Descartes built an AI agent called Debbie to solve this. She calls drivers, identifies whether the tracking app is installed, and walks them through setting it up if it isn’t. She speaks multiple languages, responds naturally, and has now completed 700,000 calls, helping over 420,000 drivers get connected.

Previously, this was handled by teams manually making calls: expensive, slow, and never fully scalable. Debbie does it continuously, and at a scale no human team could match. It’s one of the clearest examples in logistics of AI solving a problem that genuinely couldn’t be solved any other way.

The Next Step Is Autonomy

Real-time visibility was the goal for the past decade. The goal for the next one is autonomous action.

Today, when a container is delayed at port, a system can notify the relevant parties. Tomorrow, that same system will automatically rebook the onward leg, alert the leg-three carrier, and update the delivery schedule, without anyone making a single decision. The human role shifts from managing the flow to managing the exceptions, and over time, even that boundary moves.

Elmer’s vision is for one person to comfortably manage 100,000 shipments a month. That’s not a distant ambition. It’s the direction the technology is already heading.

🎧 Curious how this plays out in practice? Listen to the full Does Logistics Matter? episode with Elmer via the player on this page, or wherever you get your podcasts.

This episode is powered by Descartes

The image for this blogpost was created by AI based on the content of this podcast episode

Pin It on Pinterest

Share This