Monitoring asset performance is a crucial aspect of asset lifecycle management and maintenance strategy in asset-intensive industries. But with technologies like AI and IoT gaining traction in industrial operations, asset performance monitoring has evolved, and for good.

What is asset performance monitoring?

Asset performance monitoring is a capability that helps operators understand how well an asset performs. Asset performance monitoring is usually leveraged by asset intensive industries like manufacturing, oil and gas, cement, and power and utilities. 

In traditional setups asset performance was assessed manually, often by experts who inspected assets and collected critical data points about their performance on paper. Today, however, asset performance monitoring is done digitally, using a collection of tools, processes, and software that works together to provide insight into asset performance.

Why is asset performance monitored?

Asset performance monitoring is usually done for two reasons:

  1. Improve asset reliability and availability: By understanding how well or poorly an asset performs, operators are able to schedule maintenance and repair operations (MRO) to restore it back to its optimal functioning.
  2. Inform asset management decisions: Asset performance monitoring can also be leveraged to make better repair vs. replace decisions. It can also help uncover environmental or operational issues that lead to a suboptimal asset lifecycle.

How does asset performance monitoring work?

Asset performance monitoring is conceptually driven by 4 key processes:

  1. Capturing data points which provide insight into asset performance
  2. A software solution which provides asset performance information to operators
  3. Integration of captured data with the software solution 
  4. Using data analytics techniques to benchmark asset performance

This is enabled using various strategies depending on the type of assets whose performance will be monitored. Modern assets usually come with instrumentation capabilities built into them. They are usually also able to connect to networks using wired or wireless interfaces – and as a result, can be directly integrated into asset performance monitoring software. 

Legacy assets, however, require a different approach. Instrumentation capabilities have to be built for them. This is done using IoT (Internet of Things) sensors, which are attached to the right components, or placed at the right location to capture intended data points. They are then connected to the network, and integrated into the software solution.

Asset performance monitoring is usually not offered as a single capability. It is often a part of asset performance management or predictive maintenance platforms, and is embedded into larger processes that are driven by an asset management and risk management strategy.

What does ‘asset performance’ mean?

Asset performance may mean different things in different contexts. Asset performance is usually measured using a score – an asset health index, which tells an operator of the condition or performance of an asset based on how it compares with its historical measurements. 

In modern asset performance monitoring techniques, this score usually represents the condition of an asset at that point in time. This real-time asset performance score enables more timely maintenance, and makes it possible to mitigate failures and other risks before they occur. In leading asset performance monitoring solutions, these scores will generally be associated with an action. For example, if an asset’s performance is degrading, what is the cause, and how can it be corrected?

Building asset performance indicators

There are often multiple approaches to build asset health indicators, sometimes, even for the same asset. For example, one technique may use the same set of state variables (a set of data points, like temperature, pressure, distance between two points in a system, or vibration levels) in modeled using a different statistical technique to determine useful information like time to failure (TTF), remaining useful life (RUL), and so on.

These models may also be further tuned to better represent optimal functioning states in a given environment or process. For this, a model will be fed historical data recorded from an asset – say, a turbine generator, or an oil rig – at a particular facility over the course of years. This model is then used to create new benchmarks, against which the real-time data is compared to generate accurate asset performance indicators. Some facilities may lack useful historical records – in such scenarios, analytics solutions that can learn and self-correct themselves will prove valuable.

How does asset performance monitoring benefit asset-intensive businesses?

Asset performance monitoring, in and of itself, will only result in small, although tangible improvements in asset reliability and availability. Why? Because different teams of operators and technicians will interpret asset health or performance scores in different ways. Moreover, they will still have to make use of manual processes to understand why an asset’s performance is degrading, and how to correct it. For some machinery, this may result in downtime, because an inspection may necessitate downtime (although it can be scheduled strategically).

That’s why, asset performance monitoring unleashes its full value when it is embedded into a reliable and performant predictive or prescriptive maintenance platform. Such solutions make use of advanced analytics techniques like Machine Learning (ML) and root-cause analysis to provide actionable insights to operators. 

For instance, while asset performance monitoring will tell an operator that a generator’s performance is degrading, a predictive maintenance solution will also tell them:

  1. If the performance is degrading due to loose stator slot wedges, or damage on cooler tubes.
  2. The best action to remediate the root cause of the performance degradation.

In this manner, the corrective actions that should be taken in response to changes in asset performance are codified, and timed to extract the most value from an asset, while preventing failure and other risks.

Final words

Asset performance monitoring is a crucial aspect of modern asset maintenance and management. However, manual approaches not only do not scale well, they are also associated with high costs and suboptimal asset reliability. 

Modern approaches to asset performance monitoring based on AI and IoT make use of real-time data streams and statistical models to give real-time insights into an asset’s performance. When these insights are leveraged with a good predictive maintenance solution, they can be transformed into actions that directly result in improved asset reliability and availability. Such solutions also bring down the cost of maintenance, while directly improving the output of a facility.

Bring the power of asset performance monitoring to your asset maintenance strategy with UptimeAI – an industry-leading predictive maintenance solution. 

Contact now.

Subscribe to our newsletter