Cleaning Robots in Hospitals: Solving Staff Shortages, Reliably Meeting Hygiene Standards

At a German hospital with roughly 600 to 1,500 square meters of cleaning area per floor, an in-house cleaning team of nine works in a multi-shift schedule. The real challenge there was never the budget — it was staffing: open positions stayed unfilled for months, early and split shifts made the role unattractive to many applicants, and in this specific case the hospital's public-transport connection made recruiting even harder. That last point is an observation from this individual case, not an industry-wide finding.

Figures for the building-cleaning industry confirm the structural background: of roughly 700,000 employees, only about 35,000 are registered with the Federal Employment Agency (Bundesagentur für Arbeit) as trained specialists; more than 60% of companies report skilled-labor shortages; vacancy periods of three to eight months are common. Causes include early and split shifts, an image problem within the industry, moderate pay combined with high physical strain, and demographic change.

For hospitals specifically, the Krankenhausbarometer Facility Service 2025 (Lünendonk & Hossenfelder with KDS, surveying just over 20 hospitals) shows: 58% of hospitals rank staff shortages among the biggest challenges facing their facility services.

Add to this a structural, permanent factor: infection risk in hospitals is not an abstract regulation but a constant part of daily operations — a system that comes under additional pressure from staff shortages. The practical answer to that is not "more control," but a more targeted allocation of the staff capacity already available. This is exactly where the case described below comes in: a cleaning robot takes over the large, time-consuming public areas — corridors, reception and waiting areas, the cafeteria — freeing up time that staff need in the critical zones that still require manual disinfection.

DIN 13063 and KRINKO — A Structure the Robot Supports

Since 2021/2022, DIN 13063 has been the first German standard specifically for hospital cleaning. It builds on the KRINKO recommendation issued by Germany's Robert Koch Institute (RKI) (Bundesgesundheitsblatt, October 2022 issue) and introduces a room-group model: rooms are classified into groups by infection risk and use, each with different cleaning and disinfection requirements. The standard is not legally binding, but is recognized across the industry's trade press as an orientation framework.

This room-group model is precisely what provides the structural justification for deploying a robot: large, publicly accessible areas with a high share of floor space but a lower risk classification — corridors, waiting areas, the cafeteria — are exactly where a robot works consistently, without variation in quality. High-risk zones remain a human task and continue to require manual disinfection. The robot therefore does not replace the standard's requirements; it creates the time needed to meet them in the critical zones. Three technical characteristics support this role in practice: strict separation of clean and dirty water, increasingly automated self-cleaning of the machine itself, and seamless cloud documentation of completed tasks.

Separated Water Management as a Hygiene Advantage

Scrubber-dryer machines — whether robot-operated or manually driven — apply only clean fresh water and, in the same pass, vacuum up the contaminated dirty water; fresh and dirty water remain strictly separated inside the machine. The contrast with a classic mop-and-bucket method: there, the water becomes more contaminated with every pass while cleaning continues in the room — a well-known recontamination risk in surface cleaning. The robot used in this case operates on this scrub-and-vacuum principle and systematically avoids this risk, regardless of staff workload or how the day is going.

Self-Cleaning Maintenance Stations as an Industry Trend

An often-overlooked point concerns not the floor area but the robot itself: manufacturers are increasingly building self-cleaning functions into their systems to simplify both the machine's hygiene standard and its day-to-day maintenance — for example, automatically flushed dirty-water tanks and suction paths (Gausium Mira*), or a self-cleaning docking station introduced in 2025 for the Pudu CC1 series*, which automatically cleans the squeegee and roller brush after every run. With each robot generation, the degree of automation grows not only on the floor but also in the upkeep of the device itself — relevant for sensitive environments where the cleaning equipment itself must not become a source of contamination.

Cloud Documentation as an Automated Record

Another priority for the customer in this case was complete documentation of every cleaning task performed. The robot deployed — brand Gausium — logs every cleaning run, including completion rate, on a cloud platform. According to the manufacturer, tasks can be scheduled via app, progress tracked in real time, and detailed reports on operating metrics and historical statistics retrieved. That automated reporting is not unique to this case is confirmed by the independent fleet-management provider ToolSense, which describes "cleaning mission completion" reporting and "proof of performance" dashboards as an industry-standard feature — though without hospital-specific compliance detail.

For quality management, this is more than a convenience: DIN 13063 requires concrete testing methods (visual checks, contact-plate sampling, biological indicators) and clear assignment of tasks and responsibilities; the KRINKO recommendation calls for documented process quality. An automatically generated, gapless log replaces manual documentation with traceable, always-retrievable evidence.

Market Context: How Widespread Are Cleaning Robots in Hospitals?

Reliable, hospital-specific market data is scarce — which makes it all the more important not to mix the figures that do exist.

The Krankenhausbarometer Facility Service 2025 mentioned above (a small sample of just over 20 hospitals) shows: 19% of surveyed hospitals already use service robots, another 38% are piloting or planning deployment, and 75% name growing robot use as the most important trend of the coming years. Rising cost pressure and a tight staffing situation are cited as the main drivers — at a sample size this small, these are directional signals, not a representative market figure.

Larger, but not hospital-specific, are the figures from the International Federation of Robotics (IFR World Robotics 2025): professional cleaning robots saw 34% growth in 2024, reaching more than 25,000 units sold — the third-largest application segment in professional service robotics overall, cross-industry and not broken down by hospitals.

That Germany is also putting research weight behind the topic is shown by the RoReB project: Fraunhofer IPA, Adlatus Robotics, and InMach, together with Klinikverbund Südwest as the application partner, developed cleaning, disinfection, and automatic door-opening technologies for hospitals from 2021 to 2023, funded by the state of Baden-Württemberg. No published figures on acceptance, noise, or collision detection are available from the project — but it confirms the topic is being taken seriously beyond individual cases.

The Case in Practice

Starting Point

The case described at the outset — a hospital with 600 to 1,500 square meters of cleaning area per floor and nine cleaning staff on the in-house team — is an example of exactly this staffing problem. The robot was not introduced to replace staff, but to relieve the existing team of large-area, repetitive tasks.

A Surprise During Operation: the Patina Effect

One operational observation that was not expected in this form: on hard floors that had previously been mopped by hand for years, the first robot runs initially produced a streaky appearance — the floor looked patchier, not cleaner. The reason: over the years, a layer of grime — a "patina" — had built up that manual mopping never fully removed, because no person can apply constant, high pressure over a long period. The robot works with significantly higher and, above all, consistent contact pressure (around 50 kilograms in this case) and visibly broke down this patina after several passes — the floor looked noticeably better afterward than before the switch. This observation comes from a single case and cannot simply be generalized to every floor covering, but it shows that long-standing cleaning routines don't automatically produce the same result as a mechanically consistent process.

Three Practical Requirement Dimensions

Collision Detection

In heavily trafficked areas with patients, visitors, and trolleys, reliable obstacle detection is a baseline requirement, not a bonus feature. We've explained the underlying sensor and navigation technology in detail elsewhere (see our article on LiDAR & SLAM); in a hospital context, what matters most is that the system reacts reliably and without abrupt stops in transition areas between public traffic and narrow corridors.

Noise Level

Noise is its own requirement criterion in hospital operations — unlike, say, in an office building — especially near rest areas. In this case, the robot's factory settings (wet-cleaning travel speed, brush pressure and rotation speed) were adjusted on site. The most effective lever turned out not to be lower brush power alone, but the combination of lower brush power and faster travel speed — the robot then spends less time in front of individual doors and rooms. An informal on-site measurement (not a standardized test method) subsequently found an average of about 60 dB(A), measured directly in front of the unit — for comparison, a household vacuum cleaner can reach up to 80 dB. The figure should be understood as a practical value, not a certified measurement, but it shows that noise levels can be actively shaped during operation rather than accepted as a fixed device spec.

Psychological Acceptance

Acceptance clearly differs by group in a hospital setting — in this case, two distinct trajectories could be observed.

Patients were initially the most skeptical group, reacting mainly to the driving noise as the robot passed by in the corridor. That skepticism largely dissolved over about four days. This lines up with external, patient-focused research on the form design of service robots: one study found clearly stronger preference for non-humanoid shapes — an animal-like form with a screen scored highest at 35% approval, a cylindrical form came second at 27.8%, while humanoid forms scored lowest at 14.8%. For perceived safety, a low center of gravity (33.8%) and a soft, shock-absorbing outer material (30.7%) were rated most positively. These studies come from a different cultural context and don't transfer 1:1 to Germany, but they offer a plausible explanation for why initial patient skepticism often fades once the device's shape and behavior are perceived as non-threatening.

The cleaning team itself reacted the opposite way: positive from the start, without the job-security concerns described in some studies. The robot was perceived as relief, not as a threat — unsurprising, since the team had clearly been overloaded beforehand. One important caveat: external studies on service-robot acceptance among "medical staff" or nursing staff are not a suitable comparison here — they survey nursing or clinical staff, not the separate, in-house cleaning team that is standard practice in Germany. The cleaning team's positive reaction in this case therefore stands as an independent observation, without reference to external studies.

What This Means for Hospital Decision-Makers

The case described here shows a recurring pattern: the value doesn't lie in blanket automation of cleaning, but in deliberately taking over large, publicly accessible areas that consume a lot of time yet don't fall into DIN 13063's critical high-risk zones under the room-group model. That time saved is exactly what gives existing staff the capacity for areas that still require manual disinfection.

Three points from this case matter most for the decision: first, deploying a robot doesn't change the hygiene framework itself — DIN 13063 and KRINKO continue to apply unchanged; the robot merely supports their implementation in one part of the facility. Second, noise and driving behavior can be adjusted during operation and should be tested concretely on site before a decision is made, not judged from spec sheets alone. Third, acceptance develops differently by group and should be communicated accordingly — for patients, it's worth planning for a short adjustment period of a few days, while staff themselves tend to welcome the relief from day one, provided that benefit is clearly communicated.

Transparency note: Gausium and Pudu are part of Upgrade Robotics' distribution portfolio. Manufacturers and products are named in this article purely on editorial grounds and do not constitute a purchase recommendation.