Field and Service Robotics
Hospital Transport Robots: Where the Saving Comes From
A hospital transport robot costs 60,000 to 180,000 euro and the saving is not the porter. It is the nurse who stops leaving the ward, and that is a much larger number.
The business case is misread almost every time. A transport robot costs 60,000 to 180,000 euro and does not replace porters, who are among the cheapest staff in the building. It removes the trips that nurses make, and a nurse walking to pharmacy for 25 minutes is three to four times more expensive per hour than a porter making the same trip, and is not on the ward while doing it.
Why the porter comparison fails
A robot travelling at 1 to 1.5 m/s with a 30 to 90 second loading pause per stop is slower per delivery than a porter with a trolley who takes three items at once and knows the shortcuts. Comparing them on trips per hour makes the machine look poor, and that comparison is the wrong one for three reasons.
- Porters do more than transport. They move patients, respond to calls and handle exceptions, none of which the robot touches. Removing supply runs from a porter frees them for the work only a person can do.
- The out-of-hours trip is the expensive one. At 03:00 there is no porter on shift, so the trip is made by whoever is available, and that is clinical staff.
- Nurse trips are invisible in the labour budget. They are booked as nursing hours regardless of whether the nurse is with a patient or in a corridor, so the cost never appears as transport.
The arithmetic
| Item | Value | Note |
|---|---|---|
| Nurse trips off the ward per shift | 6 to 14 | Pharmacy, stores, laboratory, linen |
| Average round trip | 14 min | Including waiting and lift time |
| Nursing time per shift | 1.4 to 3.3 h | Away from patients |
| Shifts per year | 1,095 | Three per day |
| Nursing hours per year | 1,530 to 3,600 | |
| Share the robot can absorb | 55 to 75 % | Scheduled and routine trips |
| Hours released | 840 to 2,700 | |
| Value at 42 euro per nursing hour | 35,000 to 113,000 | Fully loaded |
| Robot cost amortised over 7 years plus service | 14,000 to 32,000 | |
| Net annual value per ward | 21,000 to 81,000 | One robot usually serves several wards |
The last line matters. Because one robot serves multiple wards, the cost line is shared while the benefit line multiplies, which is why hospital deployments are almost never a single unit.
What makes hospitals harder than warehouses
| Factor | Warehouse | Hospital |
|---|---|---|
| People present | Trained staff | Patients, visitors, children, wheelchairs |
| Lifts | Rare, dedicated | Essential, shared, busy |
| Doors | Few, controlled | Many, fire doors, access controlled |
| Floor | Flat, clear | Beds, trolleys, equipment parked anywhere |
| Route stability | High | Corridors reconfigure with the ward |
| Noise tolerance | High | Very low at night |
| Consequence of blocking | Lost throughput | A blocked corridor in an emergency |
The lift row is the one that decides most projects. A robot that cannot call a lift on its own serves one floor, and integrating with a lift controller is a building services project with its own approvals, not a software feature.
What to specify
- Lift integration, in writing, including which lift controller and who obtains the approval.
- Door and access control integration for every door on the route, including fire doors.
- Secure compartments with audit trail for controlled drugs and specimens, which is a governance requirement rather than a convenience.
- Night noise limits, measured, because a machine that wakes a ward will be switched off.
- Behaviour when blocked, since a corridor is not a warehouse aisle and waiting indefinitely is not acceptable.
- Cleaning and infection control for the surfaces, agreed with the infection prevention team before purchase.
Frequently asked questions
Do hospital transport robots replace porters?
No, and that is the wrong comparison. Porters are inexpensive and do far more than transport. The saving comes from removing the supply trips that nurses make, which cost three to four times as much per hour and take clinical staff off the ward.
What is the actual value per ward?
Roughly 21,000 to 81,000 euro net per year on a 30-bed ward, from 840 to 2,700 released nursing hours. Because one robot usually serves several wards, the cost is shared while the benefit multiplies.
What should a hospital measure before buying?
Ward exits: how many times per shift a nurse leaves the ward to fetch something, and the round-trip time including lifts and waiting. Delivery counts are the wrong metric and understate the case.
Why are hospitals harder than warehouses?
Untrained people including patients and children, shared busy lifts, many controlled and fire doors, corridors cluttered with beds and equipment, very low night noise tolerance, and a blocked corridor being a clinical risk rather than lost throughput.
What is the most common project blocker?
Lift integration. A robot that cannot call a lift serves one floor, and connecting to the lift controller is a building services project with its own approvals rather than a configuration setting.
Sources
- World Robotics 2025 report, service robotsInternational Federation of Robotics, transport and logistics segment volumes
- ISO 3691-4, safety requirements for driverless industrial trucksInternational Organization for Standardization, applies to hospital transport robots
- ISO 13482, safety requirements for personal care robotsInternational Organization for Standardization, hazards from untrained people in shared space