Author | Darren White - APM Product Manager, On Key
Reducing critical asset downtime is not primarily a monitoring problem. It is an execution problem. A business may already have sensors, inspections, alarms and an enterprise system of record, yet an asset can still fail while people compile the report, interpret the evidence, decide what matters and translate that decision into controlled maintenance work. Asset performance management (APM) creates value by protecting the last mile between warning and action.
Key takeaways
The delay that causes downtime sits between the warning and the work: interpreting evidence, deciding what matters and raising controlled maintenance work.
Condition monitoring creates the warning. Maintenance creates the outcome.
The value of APM is a shorter, more reliable route from evidence to action, not another alarm on another dashboard.
Start with the critical asset whose failure would have the most significant consequence, not with the technology.
Why does unplanned downtime persist despite condition monitoring and routine inspections?
Recognising risk before it becomes downtime
A critical asset rarely moves from healthy to failed in a single moment. Performance may drift, temperature or vibration may move beyond an expected range, energy use may rise, output may fall, or the same defect may recur. Each change is evidence. The commercial risk lies in how long it takes the organisation to recognise what that evidence means and to respond.
When the asset eventually stops, the consequence extends beyond the maintenance team. Production is interrupted. Labour and materials wait. Commitments to customers come under pressure. A short delay in deciding can become a long and expensive interruption to the operation. This is why uptime cannot be managed only at the point of failure.
Start with the reliability problem, not the data
Not every organisation starts from the same place. Some already collect large volumes of condition data but do not use it consistently. Others rely on labour-intensive routine inspections or have not yet collected the evidence needed to recognise developing failure early enough.
The common starting point is the critical-asset problem: what is undermining reliability or performance, what evidence would provide sufficient warning, and how will that warning prompt an effective maintenance action? The answer may be found in existing information or may require focused additional monitoring. In either case, the objective is not to collect more data. Improving the maintenance response is.
Condition monitoring and asset health monitoring: the warning creates the signal, maintenance creates the outcome
Condition monitoring and asset health monitoring are integral to asset performance management. They reveal what is happening to the asset. That evidence may come from existing controllers and operational systems or from additional sensors where information gaps exist. The sensor, connection and alarm are not the business outcome. They are inputs to a decision.
The key question is: what does the change mean for this asset, what are the consequences of continued deterioration, and what must happen next? Answering it requires considering condition evidence alongside asset criticality and maintenance history.
That is where APM earns its place. It helps the organisation distinguish a signal that can be observed from a condition that requires intervention and then connect that decision to controlled maintenance work. The value is not another alarm on another dashboard. It is a shorter, more reliable route from evidence to action.
When the data exists but the maintenance trail doesn’t: one critical asset, three connected pressures
Consider an operation with critical assets that are expensive to maintain, and even more expensive when they stop. Condition readings are already collected and stored, while operators monitor live dashboards and respond when alarms appear.
The reliability engineer: evidence that is hard to retrieve
The reliability engineer can see the readings but may not have a reliable record of what happened around each alarm. When the alarm clears, valuable evidence can be difficult to retrieve. Trends remain hidden, recurring conditions appear isolated, and the engineer reconstructs events from reports, spreadsheets and conversations.
The maintenance manager: delayed action
The maintenance manager faces the same gap: delayed action. A warning does not always include the context needed to determine the work, resources, and required spares. Every manual handover consumes part of the warning window.
The asset owner: lost production and rising costs
For the asset owner, the gap shows up as lost production, rising maintenance costs, and underperforming assets. The data exists, but it doesn’t consistently protect uptime or asset performance.
How On Key turns a warning into governed maintenance work
On Key closes this gap by bringing relevant condition evidence into the asset’s maintenance context. When an emerging condition requires attention, that evidence can trigger a controlled response within the On Key EAM platform.
Depending on the organisation’s agreed rules, the response may be automated, routed for investigation, or require human approval. It may include a notification or work order, predefined tasks, the appropriate resources, and required spares. The warning initiates a traceable maintenance process rather than ending an operational alarm.
The reliability engineer gains a more robust asset health history. The maintenance manager gains a reliable route from warning to work. The asset owner gains greater confidence that existing data reduces downtime exposure, controls maintenance costs, and improves critical asset performance.
Where should an APM initiative start? With the asset outcome, not the technology
A credible APM use case starts with the critical asset whose failure or underperformance would have the most significant consequence. That may be the conveyor limiting throughput, a motor without a practical standby, or a utility asset essential to the operation.
The business can then ask three disciplined questions:
Which critical failure, deterioration or efficiency loss are we trying to prevent?
What evidence do we already have, and what minimum additional evidence would give us enough warning to act?
What response and measure will demonstrate an improvement in uptime or asset efficiency?
This keeps the initiative accountable for an outcome. It also prevents organisations from mistaking the volume of connected data for progress. More readings have little value if they do not improve the timing, quality or consistency of maintenance decisions.
Example: cold-chain monitoring and electrical-current data
Cold-chain monitoring provides a practical example. Temperature records may support compliance, but combining temperature with operating measures such as electrical current can provide a richer view of how refrigeration equipment is behaving. A statistical change-detection model can analyse electrical-current data for sustained deviations from normal operating behaviour. This provides an earlier opportunity to investigate and plan work before equipment failure threatens controlled conditions or operations.
The last mile: why best-of-breed APM must integrate with your systems of record
The last mile is where best-of-breed APM must prove its value. Most asset-intensive organisations do not start with a blank technology landscape. They already have enterprise platforms for finance, procurement, materials and master data, and may also have operational technology, historians, inspection records and specialist monitoring tools. The real gap isn’t typically another data source. It is a dependable maintenance operating layer where the asset, the maintainer and the work intersect.
A best-of-breed strategy should seamlessly integrate with existing systems. The core enterprise system remains the system of record for transactions. Meanwhile, the specialised asset-management layer needs to display relevant conditions, consider asset criticality and history, trigger and oversee necessary work, support field operations, and provide useful feedback. This step is crucial to transforming asset data into actionable operational control.
This is the practical difference On Key is designed to make. On Key APM does not treat the alarm as the end of the process, and the On Key EAM platform does not isolate maintenance execution from the evidence that should inform it. Together, they provide a specialist system of action around the asset while interlocking with the organisation’s broader systems of record. The proposition is not a replacement for its own sake. It completes the chain that protects uptime.
This capability sits within an EAM platform recognised as a Leader in the Verdantix Green Quadrant: Enterprise Asset Management Software 2024.
Warning-to-work in practice: two published Martec client references
Published client references from Martec, an On Key specialist condition-monitoring partner, demonstrate both sides of this journey.
Platinum-group-metals smelter: earlier evidence protected the warning window
At a platinum-group-metals smelter, real-time partial-discharge monitoring detected developing termination defects before catastrophic failure. The warning enabled repairs to be scheduled during a planned maintenance shutdown without lost production time. Martec reports that an unplanned repair could otherwise have taken between three days and a week. Read the smelter client reference.
Manganese-ore operation: condition evidence connected to the maintenance process
At a manganese-ore operation with a single production line, Martec assessed vibration, acoustic-emission, speed, and temperature evidence remotely. Configured warnings triggered analysis, responsible personnel received notifications and recommended corrective actions, and the monitoring solution integrated with On Key EAM for maintenance planning. Martec reports improved plant availability and maintenance planning requiring fewer resources and less time. Read the mechanical plant-monitoring reference.
In the first case, earlier evidence protected the warning window. In the second, condition evidence was connected to the maintenance process needed to act on it.
What is your warning window worth?
The strongest case for APM is not built by starting with devices, protocols or dashboards. It begins with downtime the business can no longer accept, or with performance it cannot afford to leave unrealised. Choose one critical asset or asset class. Define the loss. Identify the evidence that could provide an earlier warning. Then connect that evidence to a response and measure the resulting changes.
That is how an APM conversation becomes an operational decision rather than a technology exercise. It is also how a business discovers the value already present in its asset data – provided that data reaches the right people in the right context, while there is still time to act.
Where does your warning-to-work process lose time? Book an On Key demo focused on one critical asset. We will explore how condition evidence, maintenance workflows and enterprise information in your environment can be connected to enable earlier intervention, more consistent maintenance execution and measurable improvements in uptime.
Frequently asked questions
What is asset performance management (APM)?
Asset performance management helps an organisation distinguish a signal that can be observed from a condition that requires intervention, and then connects that decision to controlled maintenance work. Its value is a shorter, more reliable route from evidence to action, not another alarm on another dashboard.
Why does unplanned downtime persist despite condition monitoring?
Because reducing downtime is primarily an execution problem, not a monitoring problem. An asset can fail while people compile the report, interpret the evidence, decide what matters and translate that decision into controlled maintenance work. APM protects the last mile between warning and action.
Where should an APM initiative start?
With the critical asset whose failure or underperformance would have the most significant consequence. Ask which failure you are trying to prevent, what evidence you already have and what minimum additional evidence would give enough warning, and what response and measure will demonstrate improved uptime or asset efficiency.
Does APM replace our enterprise system?
No. The core enterprise system remains the system of record for transactions. The specialised asset-management layer integrates with it, displaying relevant conditions, considering asset criticality and history, triggering and overseeing necessary work, supporting field operations and providing useful feedback.