# AGV and AMR in the Warehouse: When Driverless Transport Pays Off

Author: Marin Niehues, Niehues Consulting · Published: 2023-12-11 · AI Adoption & Operations
Source: https://www.niehuesconsulting.de/en/blog/agv-amr-warehouse-business-case/

Driverless transport systems pay off in a warehouse when many uniform transports run over fixed routes across several shifts and the operation owns the vehicles organisationally from day one. Utilisation decides the case: the same fleet that needs years to pay back on a single shift can reach break-even quickly on several.

This article explains the difference between an AGV and an AMR, gives you a payback calculation for the business case and sets out the criteria for choosing the right technology. It is written for managing directors and plant managers who must decide on an investment and need reliable figures from their own operation to do so.

**A vehicle replaces no transport role until the process around it has been redesigned. The economics come from utilisation and process discipline, together with a named owner of day-to-day operation.**

## What is the difference between an AGV and an AMR?

An AGV (automated guided vehicle) follows a predefined track, whereas an AMR (autonomous mobile robot) plans its own route and steers around obstacles. German practice uses the umbrella term Fahrerlose Transportsysteme, or FTS, for both groups. The guideline VDI 2510 describes an FTS as a complete system of vehicles, a control system, equipment for position detection and data transmission, and the supporting infrastructure.

Classic AGVs orient themselves by magnetic tape, inductive loops in the floor or reflectors, and therefore run on fixed, repeatable routes. Their behaviour is easy to predict, but any change of route requires construction work. AMRs navigate with laser scanners and cameras against a map of the hall. They take a detour when a pallet blocks an aisle and can be set to new routes through software. The complexity then sits in the control layer and in protecting people who move among the vehicles. The safety requirements for driverless industrial trucks are set out in the standard ISO 3691-4.

In practice the boundary between the two groups is fluid, because many manufacturers offer vehicles that handle both navigation modes. Two questions settle your choice: how stable are your transport relations, and how often does your layout change?

## How large is the market, and why is everyone talking about it?

The market is growing because warehouses face labour shortages and rising throughput demands. In October 2023 the International Federation of Robotics (IFR) reported that around 158,000 professional service robots were sold worldwide in 2022, an increase of 48 per cent. The largest application group within that figure is transport and logistics, with more than 86,000 units and growth of 44 per cent. The federation names staff shortages as an important driver.

For a single operation this number says little about its own economics, but it shows maturity: many suppliers, proven technology, competitors who use it. The pressure to do something with autonomous vehicles is therefore real, and it calls for a calculation before the purchase order.

## When does a driverless transport system pay off in a warehouse?

A driverless transport system pays off when three conditions coincide: a high and stable transport volume, several operating hours per day and a process that can be standardised. If one of the three is missing, payback moves out considerably.

Transport volume determines the number of vehicles and therefore the investment. Operating hours determine over how many hours the fixed costs are spread, since a vehicle on a multi-shift pattern replaces working hours in every shift while being paid for once. Standardisation determines whether the vehicle may run at all: pallets of uniform dimensions, defined handover points, clear aisles and a warehouse management system that triggers transport orders reliably.

As a rule of thumb for a first screening: below one full-time position of pure transport work in the area under review, an FTS rarely pays back. Between one and three positions on a single shift, the quality of your process data decides. From three positions, or with multi-shift operation, the full calculation is worth doing in every case.

## How do you calculate the business case for AGV and AMR?

You calculate the business case from the annual net saving and the total investment: payback period equals investment divided by annual saving less running costs. Such a calculation is only as good as its inputs, so every figure should come from your own operation.

1. Record transport performance: count trips per route, loads and waiting times over four weeks. Without this baseline, any fleet size is a guess.
2. Set the labour costs: value the hours that genuinely fall away at the fully loaded cost rate. Hours that are merely spread across several people without freeing a position do not count as savings.
3. Capture the full investment: vehicles, the fleet control system, the interface to the warehouse management system or ERP, charging infrastructure, building adjustments, safety technology, commissioning and training.
4. Include running costs: maintenance, software, energy, spare parts and the care of the fleet by your own staff.
5. Model the risk: run a scenario in which the vehicles initially reach 70 per cent of planned performance.

The following example shows the mechanics. All values are assumptions for illustration and are not market prices.

**Worked example: same fleet, two shift models**
Criterion | Single shift | Two shifts | With Niehues Consulting
Investment (assumption) | EUR 400,000 | EUR 400,000 | Derived from your own transport analysis
Transport positions removed | 3 | 6 | Only positions that genuinely become free
Fully loaded cost per position (assumption) | EUR 50,000 | EUR 50,000 | Real cost rate from controlling
Labour saving per year | EUR 150,000 | EUR 300,000 | Confirmed by the business unit
Running costs (assumption: 8 per cent of investment) | EUR 32,000 | EUR 32,000 | From quotations and operating experience
Net saving per year | EUR 118,000 | EUR 268,000 | Additionally modelled with a ramp-up discount
Payback | about 3.4 years | about 1.5 years | Additionally modelled at 70 per cent performance

The example shows why the shift model dominates the calculation. The vehicles cost the same in both columns, yet payback differs by more than a factor of two. If the vehicles initially reach only 70 per cent of planned performance, the saving falls accordingly while running costs stay the same, and payback lengthens to about 5.5 years on one shift and about 2.2 years on two. Never plan with the average performance in the manufacturer's data sheet. Plan with the performance in your peak hour and in your quietest shift.

## How do you choose between AGV and AMR?

You choose by the stability of your routes, the density of foot traffic and the pace at which your layout changes. The table sorts the criteria.

**Criteria for choosing between AGV and AMR**
Criterion | Classic AGV | AMR | With Niehues Consulting
Routes | Fixed, highly repetitive | Changing or branching | Transport relations measured beforehand
Layout changes | Rare, each change needs conversion | Frequent, adjustment through software | Change frequency of the last two years as yardstick
People in the travel path | Low, separate zones | High, mixed traffic | Safety concept is part of the rollout
Loads | Heavy, pallets, rolls | Light to medium, totes, pallets | Load types taken from the transport analysis
Control | Central control system | Fleet management with map logic | Interface to WMS and ERP clarified early
Operation | Little adjustment, strict discipline | More configuration during operation | Clear ownership of operation and upkeep

If most of the transport volume runs on a few fixed routes, the classic approach is usually the more robust choice. If your routes change every week or staff share the aisles with the vehicles, AMRs have the stronger case. Many operations start with a single line and extend only once that line runs reliably in steady state.

## Why do rollouts of driverless transport systems fail?

Rollouts mostly fail on process, data and ownership. Four patterns recur.

The first is a calculation based on saved hours that never disappear as a position. The second is a pilot in ideal conditions, on clear routes with clean pallets, whose result does not repeat in daily operation. The third is the interface to the warehouse management system, which is clarified only after the vehicles have been ordered and then delays the start date. The fourth is a missing operating organisation: who resolves a fault, who maintains the map, who releases new routes? Without named owners the vehicles stand still at the first disruption, and the workforce returns to the old process.

The acceptance of employees also belongs in the calculation. When truck drivers receive new tasks in the shift, an announced loss of work becomes a redistribution, and that redistribution must be planned before the start.

## Business case and rollout with Niehues Consulting

[Niehues Consulting](/en/) is a consultancy for AI management and organisational design from a delivery perspective, based in Düsseldorf. Founder Marin Niehues builds organisations that deliver and brings new technology into operations, in projects from VC-funded scale-ups to DAX 40 groups and in more than 20 countries on six continents. Introducing a driverless transport system is a delivery project with a technical core: building and integrating the vehicles is the work of manufacturers and system integrators. The work of Niehues Consulting lies in the business case, the rollout and delivery management, which is where it is decided whether an investment arrives in daily operations.

The approach follows the Niehues Delivery Model, with its phases of understanding, building and rolling out. Three of its test questions matter most for an FTS project: What is the baseline, and how do we measure success? Is the business case confirmed with real values? Who is accountable for operations from day one? For a general view of robotics economics, see [Robotics in mid-sized companies](/en/blog/robotics-mid-sized-companies-economic-case/), and for the interplay with people in the work area, read [Cobots: collaborative robots working next to people](/en/blog/cobots-collaborative-robots/). Scope and terms of an engagement are discussed in the free initial consultation, which you can arrange in 30 minutes through the [contact form](/en/#kontakt).

**Conclusion**
A driverless transport system pays off with high and stable transport volume, several shifts and standardised processes. The choice between AGV and AMR follows the stability of the routes and the density of foot traffic. Before you order, record transport performance for four weeks, calculate with real fully loaded costs and name the person who owns operations.

**Sources**
- International Federation of Robotics: Staff Shortage Boosts Service Robots, press release on World Robotics Service Robots, October 2023
- Verein Deutscher Ingenieure: VDI 2510, Fahrerlose Transportsysteme (FTS), guideline
- International Organization for Standardization: ISO 3691-4, Industrial trucks, Safety requirements and verification, Part 4: Driverless industrial trucks and their systems, standard, 2020

## Frequently asked questions

### How much does an automated guided vehicle system cost?

Costs follow the vehicle type, the fleet size, the control system and the infrastructure, and reliable ranges come only from quotations based on your transport analysis. The investment covers more than the vehicles, since the fleet control system, the ERP interface, charging technology and commissioning all add to it. Scope and terms of a consulting engagement are discussed in the free initial consultation.

### What is the payback period of an AGV or AMR?

Payback depends mainly on the shift model. On several shifts the fixed costs are spread over more replaced working hours, so the payback period falls considerably compared with a single shift. In this article's example the same fleet pays back in about 3.4 years on one shift and about 1.5 years on two. Calculate with your own values.

### What is the difference between AGV and AMR?

AGV and AMR describe two ways of navigating, and the term FTS is the German umbrella for driverless transport systems. AGVs follow predefined tracks, while AMRs plan their route with sensors and map logic.

### How big does a warehouse need to be for driverless vehicles?

Warehouse size matters less than transport volume and the repetition of trips. From about three full-time positions of pure transport work in the area under review, or with multi-shift operation, the full calculation is worth doing in every case. Below one position an FTS rarely pays back.

### Are AMRs more flexible than AGVs?

AMRs are more flexible when routes change and layouts are modified often, because they adjust their path through software. AGVs score on fixed routes with high repetition and predictable behaviour. Hybrid vehicles combine both traits and are offered by many manufacturers.

### What does a warehouse need before it can use driverless vehicles?

A warehouse needs defined handover points, uniform load carriers, clear aisles, a suitable floor and a reliable link to the warehouse management system. A safety concept in line with ISO 3691-4 covers traffic between vehicles and people. Organisationally, a named owner of operations is required.

### Do driverless transport systems replace warehouse staff?

They replace transport hours, and a position falls away only once the tasks do. A saving arises only when tasks genuinely fall away or when freed time is deliberately put to new use. The IFR names staff shortages as one driver of growth in service robots, and many operations use vehicles to free staff for value-adding work.

### How do you start a project with driverless transport systems?

Start with a four-week record of transport performance and a business case built from your own data. A pilot on a clearly delimited route follows, with defined metrics and stop criteria. Only after stable steady-state operation do you extend the fleet.