Autonomous Mobile Robots: Optimising Internal Logistics

The hours nobody counts

In most plants and warehouses there's a cost that never shows up on any dashboard: skilled people pushing carts. Moving material from the warehouse to the lines, clearing finished product, shifting pallets between processes. Thousands of hours a year that add no value to the product, yet consume some of the hardest staff to hire. Autonomous mobile robots — AMRs — exist precisely to recover those hours.

How an AMR works

An AMR navigates the plant autonomously without any physical infrastructure — no rails, no magnetic tape, no reflectors. Using laser navigation and simultaneous localisation and mapping (SLAM) algorithms, the robot builds a virtual map of its surroundings, always knows where it is, and plans its routes in real time. If it encounters an obstacle — a person, a badly parked pallet, a cart in the way — it detects it and goes around. It doesn't stop and wait for someone to move it: it makes a decision and carries on.

This is the fundamental difference from traditional AGVs, which follow fixed routes defined by physical guides. An AGV is robust in environments where nothing changes; the problem comes when something does change, because modifying its route means modifying the installation itself. With an AMR, changing the route just means updating a map in software.

Where it delivers the most: common applications

Transport between warehouse and lines

The most widespread use case: the AMR brings raw material to the lines and clears finished product, receiving missions from push-button stations, tablets, or directly from the plant's ERP or MES. The production order triggers the transport without any human intervention.

Cyclical routes: the automated milk run

Periodic runs that visit several stations, picking up and dropping off material at set intervals. It's the robotic version of the manual logistics train, with one difference: punctuality no longer depends on who's on shift that day.

End-of-line clearing

Combined with a palletizing robot, the AMR removes the finished pallet and brings an empty one. The result is a fully autonomous end of line — one of the fastest-growing configurations in industry.

Loads of every size

The current range covers everything from agile ~300 kg platforms — built for narrow aisles and variable flows — to units capable of moving 1,000 kg or more, eliminating repetitive pallet-truck and forklift runs. Manufacturers such as Youibot have developed complete families for each of these ranges. And there's a side benefit rarely factored in: forklifts are responsible for a significant share of plant accidents. AMRs have no blind spots and never exceed safe speeds around people.

The fleet manages itself (almost)

AMRs run under fleet-management software that assigns missions, optimises routes, manages priorities and sends each robot to charge automatically. Integration with ERP, MES or WMS via API lets missions generate themselves and the system report status and traceability for every movement. That traceability, in turn, generates data: which routes get congested, which processes wait on material, where the real bottleneck in intralogistics actually sits.

When investing makes sense, and when it doesn't

The starting calculation is simple: how many hours a day the plant spends on repetitive internal transport. As a general rule, from 4-6 hours a day upward, an AMR working up to three shifts pays for itself in under two years, often well before. Factors that speed up the return:

• High-volume, repetitive flows. There's no return on a robot making two trips an hour.

  
  

• Extended shifts or weekend operation, where adding logistics headcount isn't viable.

  
  

• Layouts that change frequently, because software-based reconfiguration avoids the cost of modifying installations.

And let's be honest about the other side: in small facilities with very irregular, unpredictable flows, manual transport can still be the more efficient option in the short term. Automation for its own sake isn't the goal.

How to get started without getting it wrong

The sensible path has three steps. First, map the flows: origin, destination, frequency and load type for every repetitive route. Second, run a pilot with one flow and one robot: mapping the plant and commissioning an AMR can be completed in days, letting you validate the business case with minimal risk. Third, scale the fleet incrementally from the same management software.

Internal logistics is one of the largest pools of untapped productivity left in most plants, precisely because nobody measures it. AMRs are the fastest, least disruptive way to unlock it: no construction work, no infrastructure, and a return you can measure from month one. The difference between an AMR project that transforms a facility and one that falls short almost always comes down to the upfront flow analysis. The robot is the easy part.