What is the difference between condition-based and predictive maintenance?

Condition-based maintenance acts when a measured signal crosses a limit now. Predictive maintenance uses trends in those same signals to forecast when a failure is likely, so work can be planned earlier.

What sensors does condition-based maintenance need?

It depends on the failure modes you care about. Vibration sensors suit motors and pumps, temperature and pressure suit compressors and coils, and differential pressure suits filters. Many signals already exist in a BMS.

Does condition-based maintenance replace preventive maintenance?

Usually it reduces it rather than removing it. Statutory inspections and safety checks still follow their schedules, while condition data lets teams skip or delay work that the asset does not yet need.

Can I start condition-based maintenance on a few assets?

Yes. Most teams start with critical assets where a failure is expensive, such as chillers, air handlers or process pumps, then extend the approach as limits and responses mature.

Glossary

Operations and asset management

What is Condition-based maintenance?

Definition

Condition-based maintenance is a strategy that triggers maintenance work when measured equipment condition, such as vibration, temperature, pressure or current draw, crosses a defined limit, rather than on a fixed calendar. It reduces unnecessary servicing and catches deterioration earlier, and it supplies the measurement layer that predictive maintenance builds on.

What it is

Condition-based maintenance, often shortened to CBM, replaces fixed service intervals with measured evidence. Instead of changing a bearing or cleaning a coil every so many months, the team sets limits on the signals that show wear and acts when a signal crosses its limit.

Typical signals include vibration on motors and pumps, discharge temperature on compressors, differential pressure across filters and strainers, current draw on fans, and run hours. Each one is tied to a failure mode the team already understands.

CBM sits between reactive maintenance, which waits for a failure, and predictive maintenance, which forecasts when a failure is likely. CBM answers the question: is this asset in a condition that needs work now.

Where it is used

CBM is common wherever rotating equipment and fluid systems matter: hospitals, airports, data centers, manufacturing plants, dairy processing, campuses and commercial portfolios. It suits assets where a failure is costly and where wear shows up in a measurable signal before the asset stops.

How it is measured

A CBM program needs a sensor or meter on each monitored asset, a defined limit for each signal, and a record of what was done when the limit was crossed.

Vibration, temperature, pressure and current readings per asset
Limits set from manufacturer data or the asset's own history
Alarm history showing when each limit was crossed
Work orders and closure notes linked to each alarm

How Aqueduct helps

Aqueduct reads the condition signals your equipment already produces, or new sensors installed through partners, and turns a crossed limit into a routed fix.

Per-asset baselines for vibration, temperature, pressure and run hours, refreshed every 60 seconds
aQ PredictIQ flags condition drift and explains the likely cause in plain language
Tickets go into your CMMS with a recommended fix; Aqueduct does not replace your CMMS
After the work, the same signal confirms the asset is back within its limit