Maintenance & Uptime

Preventive and predictive maintenance

Robots need regular care like any other equipment. Preventive maintenance follows a schedule: inspecting joints and cables, replacing wear parts such as grippers or wheels, checking batteries and sensors, and cleaning cameras. Predictive maintenance uses data the robot already records, such as motor currents, temperatures and error logs, to spot problems before they cause a stoppage. Many modern robots support remote monitoring, which lets the maker or your own team watch fleet health from one place. Ask each maker for the recommended maintenance schedule, and which tasks your staff can safely do themselves.

Spares, repairs and response

When a robot stops, what matters is how quickly it works again. Ask which parts are most likely to need replacing, whether spares can be held on site, and how long parts typically take to arrive. Find out who carries out repairs: the maker, a local service partner or your own trained technicians. For critical processes, consider how the work continues while a robot is down, whether through a spare unit, a manual fallback or rebalancing work across the fleet. Settle these answers before go-live, not during the first breakdown.

Measuring uptime properly

Agree how uptime is defined before you measure it. Does it count only unplanned stoppages, or planned maintenance too? Does a robot that is running but needs frequent help count as available? Useful measures include availability during scheduled hours, mean time between failures, mean time to repair and interventions per shift. Track them per robot and per task, and review them regularly with the maker or service provider. If you have a service agreement, make sure its definitions match the ones you use internally, so both sides are measuring the same thing.

Sources and further reading

• ISO 22400-2:2014 Key performance indicators for manufacturing operations management — International Organization for Standardization.
  Defines manufacturing KPIs and their formulas, a basis for agreeing how uptime is measured.
• ISO 14118:2017 Safety of machinery: Prevention of unexpected start-up — International Organization for Standardization.
  Requirements for stopping machinery from starting unexpectedly while people work on it.
• ISO 55000:2024 Asset management: Vocabulary, overview and principles — International Organization for Standardization.
  Principles for managing physical assets, such as a robot fleet, across their whole life.

Common questions

Can our own staff maintain the robots?

Often, for routine tasks such as cleaning, inspections and swapping wear parts. Ask each maker which tasks are user-serviceable and what training is needed.

What does mean time to repair measure?

The average time from a robot stopping to it working again. Read alongside mean time between failures, it shows how reliable a fleet is in practice.