Rearranging an existing plant for flow: what it really costs to move a machine

Trial it in tape and cardboard before booking the crane

A rearrangement drawn on a plan looks convincing and hides the things that make it unworkable: the operator cannot see the second machine from the first, the pallet truck cannot turn, a door swings into the walkway, the material presentation forces a twist that will not survive a full shift. Marking the proposed arrangement on the floor with tape, using mock-ups for footprints and walking the cycle with the people who will run it finds these in an afternoon. It also converts scepticism into participation, because operators who would have resisted the move end up altering it.

Services, foundations and extraction decide what is movable

The machine is often the easy part. Electrical supply, compressed air, cooling water, drains, dust or fume extraction, foundations for anything that vibrates, overhead crane coverage and floor loading all constrain where equipment can go, and relocating them commonly costs more than the lift itself. Bring the plant engineer in at the sketch stage rather than after the layout is agreed. It is also normal for the survey to establish that one machine simply cannot move, in which case the flow design has to be built around it instead of pretending otherwise.

A cell is an organisational change wearing a physical disguise

Putting the machines together achieves nothing unless the people, the scheduling and the ownership change alongside them. A cell needs its own team able to run several of the operations, a single point at which work is released to it, its own measures of output and quality, and a supervisor accountable for the whole cell rather than for one machine type. Plants that move the equipment while keeping departmental supervision, separate operator groups and machine-level scheduling end up with the same queues in a more expensive arrangement, and then conclude that cells do not work here.

Disruption is the budget line that gets underestimated

Moving equipment means production lost during the move, a period of unstable output while operators relearn, requalification of the process where a customer or regulator requires it, recalibration, and the real possibility that an old machine never runs quite the same again. The plan needs a stock build to cover the outage, a sequence keeping one route running while its neighbour moves, and an explicit decision about which orders are at risk. Staging the work across quieter periods usually beats one ambitious shutdown, even though it takes longer overall and looks less decisive.

Safety and access are re-approved, not inherited

A machine that was safe in its old position is not automatically safe in its new one. Guarding, emergency stops, isolation points, escape routes, pedestrian and vehicle segregation, lighting, noise exposure at neighbouring workstations and access for maintenance all need reassessing, and in many jurisdictions the risk assessment must be revisited before the equipment is operated again. This is not a formality: rearrangements that squeeze equipment closer together typically shrink maintenance access and walkways first, and the consequences appear later as poor maintenance and near misses rather than immediately.

Frequently asked questions

Do we have to move machines to improve flow?

Often not, and it should be among the last options tried. Reducing batch sizes, releasing work differently, changing the transport method between steps, staging material closer to the point of use and improving changeovers all shorten elapsed time without touching foundations. Move equipment when distance and handling genuinely prevent the flow you want, when the same operator needs to run consecutive operations, or when the transport itself is causing damage and delay.

How do we handle a machine that several product families need?

Either leave it where it is and treat it as a shared service with a controlled buffer, or duplicate it if a smaller and less expensive unit can serve one family on its own. Placing a shared machine inside one cell gives that cell priority and starves the others, which surfaces as a scheduling argument rather than a layout complaint. Where duplication is affordable, a second smaller machine dedicated to a family is frequently better value than a faster shared one.

When is rearrangement not worth doing?

When the product range is about to change, when the building is due to be vacated, when the equipment is close to replacement, or when the flow problem is really a scheduling and batching problem the move will not touch. Also when the site cannot afford the outage during the period proposed. A rearrangement is a long-lived commitment made with imperfect information, so it should follow a flow design the plant has already tested by cheaper means.