For airport and rail terminals, specify a skylight cleaning robot that runs below 65 dB, works on battery without a live cable over public space, and cleans 700+ m2 per night shift. Lingkong K3-class machines fit these roofs; light residential units do not.
Terminals combine every constraint at once. Up to 15,000 m2 of glass, four-hour access windows, a public concourse underneath, and acoustic limits from local authorities. Sites that buy a general-purpose machine for this job regret it within a month.
Why are transport terminals harder than shopping malls?
Three reasons. Size: a major rail terminal canopy can present 6,000 to 12,000 m2 of cleanable glazing in one span. Access: work usually happens between 01:00 and 05:00, so setup speed decides how much of the shift is productive. Public below: nobody wants a machine, a drip tray or a bucket landing in a departure hall.
Add structural depth. Terminal roofs are deep trusses with sensors, sprinkler lines and vents everywhere. The machine has to be lifted over and around obstructions without losing time on every one.
What noise level is acceptable on a night shift?
Most transport operators target under 65 dB at 10 metres and under 55 dB in adjacent residential zones. Suction-based glass cleaning robots generally sit in the 60 to 70 dB range at close distance, which is why many terminals run them on the far span first and finish near the perimeter after the last service.
| Constraint | Typical target | Implication |
|---|---|---|
| Access window | 4 hours per night | Fast setup, battery operation |
| Noise at 10 m | Under 65 dB | Rules out older high-vacuum units |
| Water spill risk | Sealed loop preferred | No open buckets over public space |
| Coverage per night | 700-1,000 m2 | Two machines for large canopies |
How many machines does a terminal actually need?
Run the arithmetic on real coverage, not pass rate. A canopy of 8,000 m2 cleaned four times a year over four-night windows needs two machines working the full window. One machine means six to eight nights, which operations rarely approve.
Two machines also give redundancy. If one develops a pad fault at 02:00, the shift is not lost. On transport work that redundancy is worth more than the throughput.
What else should be in the specification?
- Vacuum logging so each pass is traceable after an incident report.
- Tether points rated for the machine mass plus margin.
- Battery operation, with no live mains cable crossing public space.
- Compact enough to pass through service corridors and lifts.
- Remote or semi-autonomous return-to-dock to cut repositioning time.
That last point is underrated. On a huge canopy, walking the machine back to a dock every 40 minutes eats the shift. Machines that can return themselves save 30 to 45 minutes per night.
Who should not use robots here
Terminals with listed or heritage roofs, or with integrated photovoltaic glazing where a robot’s weight and pad contact could stress thin-film modules, should not put a machine on the roof without a structural review. And if the operator cannot guarantee a night window, robots will not help — they still need a controlled space.
If you manage a terminal roof and want a realistic plan, send the drawings through our contact page. The malls, airports and factories case review covers what these sites typically end up buying.
How does weather interrupt a terminal schedule?
Badly, and often. Wind above 25 km/h stops work on exposed canopies, and summer storm cells can roll in within the four-hour window. Build one contingency night into every cycle. Sites that plan for exactly four nights get three productive ones; sites that plan five get four and stop arguing with the weather.
How much does a night shift really cost?
Add the access premium, a supervisor, transport and the recovery day for staff. Night work carries a wage premium in most regions, and the productivity per hour is lower. That is why terminals measure cost per square metre per year across the whole cleaning programme rather than per visit. The comparison only makes sense at the programme level.
What does a typical cleaning cycle look like?
A full cycle is more than the time the machine spends cleaning. It is the walk to the roof, the water fill, the boundary setup, the run itself, and the pack-down. On a well-mapped roof the setup is 20 minutes and the run dominates. On a difficult roof with restricted access it can be 45 minutes before the machine lays down its first pass. Budget the whole cycle, not the pass, when you plan a shift.
How does glass type change the brush and pad choice?
Coated glass, laminated units and textured or fritted panels all behave differently under a rotating brush. Fritted glass wears pads faster and needs lower brush pressure. Coated low-E glass scratches easily, so a clean pad and a fine brush are not optional. Textured or prismatic glazing traps dirt in the pattern and often needs more passes, which quietly doubles your clean time on those sections.
What training does an operator actually need?
Half a day is enough for basic operation, but competence comes from the third and fourth cycle. Train on the boundary rules, the vacuum alarm, the recovery procedure and the pad inspection, and record it. The single most valuable habit is the pre-cycle pad check, because a dusty or torn pad skirt is the most common cause of both sliding and scratching. Make it a signed checklist, not a habit in someone’s head.
Key Takeaways
- Terminal roofs combine scale, night-only access and public space below.
- Target under 65 dB at 10 m and battery-only operation over public areas.
- A 700-1,000 m2 per machine per night figure is realistic, not 2,000.
- Two machines give redundancy that matters more than raw throughput.
- Self-returning machines save 30-45 minutes of repositioning per shift.