The end of the forklift era in production? How to phase them out safely

More and more plants are taking a decision that would have sounded bold a decade ago: eliminating forklifts from production. This is not about fashion or a crusade against forklifts — it is about a calculation of safety, cost and space which, in the assembly area, now weighs against on-demand transport. In this article, we explain why forklifts are being phased out of production halls, where they still make sense, and how to make the transition in stages without transferring the chaos to a larger means of transport.

Why the forklift is disappearing from the production area

A forklift is irreplaceable where the load is heavy, uniform and moved infrequently. The problem starts when the same vehicle serves an assembly line — an environment with many pedestrians, narrow aisles and frequent, small deliveries of components.

Three reasons dominate conversations on the shop floor:

• Safety. A forklift in production means highly variable traffic — it drives "on call", along a changing route, at a speed dictated by how rushed the operator is. A tugger train runs a fixed loop, at low speed, along the same route. Less variability means fewer points of collision with pedestrian traffic.
• Cost. A well-calculated system makes it possible to replace 3–4 forklifts with a single tugger train. That covers not only the fleet, but also fuel, servicing and operators.
• Space and inventory. Reactive transport forces a buffer at the line that grows "just in case". A fixed loop allows this buffer to be reduced to a level calculated from actual consumption.

A key caveat: if deliveries today work "on call", there is not yet a process to automate. There is a series of reactions to shortages. Phasing out the forklift starts with organising the flow, not with buying a vehicle.

Raw material versus finished goods — not everything travels in the loop

The first distinction to make concerns the direction of the flow. A milk run — a fixed delivery loop by tugger train along a set route, to the same points, at a fixed interval — is well suited to supplying the line with components: KLT containers, small parts, pallet units exchanged on a "one full in, one empty out" basis. Empty packaging returns in the same cycle instead of requiring a separate trip.

The flow of finished goods and bulky waste looks different. Pallets of finished product heading for the docks, heavy units for high-bay storage, unusual dimensions — these are tasks for which a forklift or dedicated transport is often the right choice. It makes sense to separate these two worlds: the milk run loop supplies production, while the forklift serves high-bay storage and shipping. Trying to squeeze everything into a single loop ends with an overloaded train and delays.

Real implementation barriers that presentations do not mention

Phasing forklifts out of production most often founders not on technology but on the geometry of the hall. Four barriers come up regularly:

• Dead ends in the layout. The loop must close. A route that ends at a dead end forces the train to turn around or reverse — and that is the riskiest manoeuvre for a train of trailers.
• Width of transport routes. The CAD drawing almost always lies. The real width of the route has to be measured on the shop floor (Gemba Walk) — with pallets parked by the line, buffers and people in the aisle. A tractor with four trailers needs a verified turning radius at every bend, not a declared one.
• Heavy loads. Standard trailer platforms have their limit. Transporting loads over 500 kg requires a reinforced chassis with all wheels steered by the drawbar — otherwise tracking and safety suffer. This is the threshold at which the design has to be chosen deliberately, rather than "adding it to an existing trolley".
• Narrow production cells. Last-metre delivery into a tight cell requires trolleys that can manoeuvre in a small space. Solutions such as modular CS system trolleys with a diamond wheel arrangement and 360° rotation were designed precisely to bring a structure into a very narrow cell without repeated manoeuvring.

Each of these barriers can be overcome — provided you measure the load carrier, the exchange time and the geometry of the station before you choose the equipment. Equipment is selected last, starting from the process, not from the catalogue.

Anatomy of a train — detachable trolleys versus the mother–daughter system

A tugger train is not a "vehicle" but an entire material exchange system: the tractor, the frames or trailers, the load carrier, the supermarket and the exchange standard at the station. At the heart of the decision is how the full carrier is set down and the empty one picked up.

In practice, there are two approaches:

• Detachable frames (E-frame / C-frame). The platform is rolled into the trailer from floor level, with a quick "hot swap" exchange. They provide stable tracking and low operating costs. The limitation of some of them is that they are one-sided — delivering to one side of the aisle only. They also need lateral space for the exchange.
• Mother–daughter system. This works well where there are many short stops. It offers the fastest exchange at the station, at the price of a higher cost for the "daughter" trolleys. The more stops on the loop, the more the speed of exchange counts — and the more this approach pays off.

The choice of frame follows from measurement: the dimensions of the load carrier, the time needed to set down the full one and pick up the empty one, the space at the station, the turning radius and the number of stops. Only these data point to the technology.

Eliminating forklifts from production step by step: from milk run to AGV

The most common mistake is jumping from forklifts straight to robots. The migration path runs through five levels of logistics maturity, and none of them may be skipped:

1. Chaos. Forklifts on demand, no data, no standard. First, designate set-down areas, markings and order.
2. Order. Fixed delivery points, 5S, a full/empty standard — but deliveries are still reactive.
3. Manual milk run. An operator with trolleys on a fixed route. This is where you set the logistics takt and measure real times.
4. Tugger train. A tractor with frames, the supermarket as the starting point of the loop, standardised load carriers and punctual departures.
5. AGV. Automation of a stable, predictable loop.

The rule is firm: without a stable manual milk run, there is no talk of automation. AGV/AMR robots do not fix a process — they replicate it. Automating an unstable loop is the most expensive mistake possible. A practical threshold: the manual train should run punctually (departure deviation of less than 2 minutes) for at least four weeks before you even raise the subject of navigation — magnetic, laser or 3D LiDAR.

What about loads that will not travel by milk run

Not every material will get on the train. Cable harnesses, panels transported upright, polystyrene and fabric waste, parts with delicate surfaces, reels, loads of extreme dimensions — these are tasks for manual transport: dedicated workstation trolleys, FIFO gravity racks and solutions that reduce lifting. Here, a manually guided trolley tailored to a specific part is often more ergonomic and safer than any loop.

And an important final point: the forklift does not disappear from the plant. It stays where it makes sense — at the docks, for unloading deliveries and in high-bay storage. The aim is not to wipe the forklift off the map, but to move it out of the production area, where it generates the most risk. Wherever people and vehicles share space, pedestrian protection remains the complementary layer — barriers, fencing and zone marking.

Where to start — a practical checklist

• Separate the flows: what travels in the loop (line supply) and what stays with the forklift (finished goods, high-bay storage).
• Walk the route physically and measure the real route widths and turning radii — with pallets and people, not from the CAD drawing.
• Measure the load carrier, the weight of a full container and the exchange time at the station before choosing the frame type.
• Mark the heavy-load threshold (over 500 kg) and plan a reinforced chassis or separate transport for such loads.
• Stabilise the manual milk run before calculating the return on an AGV.
• Secure the zones shared by pedestrians and vehicles.

Does a milk run always replace all forklifts?

No. A milk run replaces on-demand transport in the production and line-supply area. Forklifts remain at the docks, for unloading and in high-bay storage, where their lifting capacity and vertical reach cannot be replaced.

What is the real replacement ratio?

A well-calculated tugger train replaces 3–4 forklifts. A poorly calculated one merely transfers the chaos to a larger means of transport, which is why it is crucial to measure the logistics takt and the loop cycle time beforehand.

When should you move to AGVs?

Only when the manual loop runs predictably: punctual departures, standardised load carriers, a stable supermarket and a loop that does not require emergency support from a forklift. Without this, automation magnifies existing problems instead of solving them.