Predictive vs Condition-Based Maintenance: Which is Best for Me?

Staying on top of your duties and meeting your compliance responsibilities as a facility manager requires sound knowledge of your building and the right maintenance strategies for your organisation.  

At SFG20, the industry standard for building maintenance, our whole purpose is to help you uphold the highest possible standard of building safety. With over 30 years of industry experience, we know just how vital it is to find the right maintenance strategies to prevent building safety risks and achieve compliance.   

To help you make an informed decision, let’s explore the differences between predictive and condition-based maintenance, two common but distinct proactive maintenance strategies. You’ll gain an understanding of when and how you’d apply each strategy, as well as see how the pros and cons for each one measure up. 

What is Predictive Maintenance?

Predictive maintenance is a proactive strategy which helps to maintain and prolong the lifespan of assets, resources and infrastructure. 

Often referred to as PdM, it uses real-time, condition-based monitoring to identify potential issues before an asset breaks down and combines historic data analytics and machine learning to forecast future failures. 

The goal of this strategy is to reduce unplanned failures and the associated maintenance costs. 

This information is then combined with historical data which can come from in-house knowledge or collection sources from further afield. 

PdM is typically used for specialised equipment that has: 

• Failure modes that can be monitored effectively
• An essential operational function

PdM uses a number of techniques for predicting and preventing equipment failures, including: 

Vibration Analysis 

By collecting vibration signals through specialised sensors, a machine’s patterns can be recognised and compared to real-time data. 

This helps to detect when a machine is out of sync, which can then indicate potential failures before they become a problem. 

Vibration analysis is primarily used on equipment with rotating components that use bearings, such as:  

• Motors 
• Pumps
• Fans
• Conveyors
• Turbines
• Generators

Acoustic Analysis 

By monitoring sound frequencies on operating machinery, acoustic sensors can help to detect changes in sound patterns. 

This can point to issues such as friction and stress, which could indicate deterioration. Rotating machinery in particular benefits from this type of technique. 

Acoustic analysis can also be used for processes involving fluid flow in pipes and pressure vessels. 

Oil Analysis 

Any machinery that requires oil must have an oil analysis process.  

This helps to: 

• Improve performance
• Avoid malfunctions
• Increase lifespan

This process involves measuring properties such as temperature and viscosity and looking for foreign bodies. 

It’s worth knowing that while elements of this can be carried out in-house with electronic sensors or specialist equipment, you may need to collect and send samples to be analysed by a qualified laboratory professional. 

Infrared and Thermal Monitoring 

Temperature monitoring can be used to look for issues in machinery, electrical systems and even building structures.  

These sensors are used to detect abnormal heat signatures, which can point to potential failures before they happen. 

The Pros and Cons of Predictive Maintenance 

The Pros of Predictive Maintenance  

• Saves you money by reducing unnecessary repairs and replacements. 
• Prevents losing valuable time on unexpected breakdowns or carrying out unnecessary maintenance work. 
• Ensures assets are running properly to create a safe working environment for building occupants.
• Increases asset lifespan.

The Cons of Predictive Maintenance  

• Investing in these types of sensor systems can be costly to start with.
• Businesses that don’t have experience with these systems may find it complicated.
• It can take some time to gather enough data to get predictions from the system.
• As with all technology, this type of maintenance strategy is not inerrant, whether that’s data that hasn’t been recorded properly or a sensor that’s been disconnected, all of which affecting accuracy.

What is Condition-Based Maintenance?

Condition-based maintenance (CBM) is a proactive strategy which uses sensors to ensure that assets are monitored in real-time, and that maintenance is only performed when declining performance levels have been identified.  

CBM is typically used on assets that have an essential operational function and are likely to result in higher costs should they break down unexpectedly. 

There are several common CBM techniques, which include: 

Acoustic Analysis 

Acoustic analysis is commonly used for machinery that uses motors, pumps and compressors. 

Sensors monitor sound patterns and pitch to determine if a machine is running efficiently.  

Any changes can indicate issues such as bearing failure or loose connections. 

Vibration Analysis 

Monitoring vibration patterns can detect if a machine is out of sync and therefore not running as efficiently as it should. 

This can help to detect possible failures in machinery with rotating components that use bearings.  

These parts create a certain level of vibration, which increases as they degrade or misalign.

Ultrasonic Analysis  

Using high-frequency sound waves, ultrasonic analysis detects deep flaws in materials.  

These flaws can lead to leaks or faults in electrical systems, or catastrophic failure in rotating components that use bearings.   

Infrared and thermal monitoring  

A safe and efficient non-contact strategy, infrared thermography creates visual images of temperature distribution to detect abnormal heat signatures such as hotspots or fluctuations.  

Oil Analysis 

Assessing the oil in a machine can help to determine wear and tear, which can aid in diagnosing specific issues before they lead to failure. 

The Pros and Cons of Condition-based Maintenance  

The Pros of Condition-based Maintenance  

• Reduces costs by acting on potential failures before they become a bigger and more expensive issue.
• Only targets assets that need maintaining.
• Ensures assets run smoothly and reliably.
• Helps to avoid losing valuable time on unexpected breakdowns.
• Minimises disruption. 
• Creates a safer environment for workers by ensuring equipment is running properly.

The Cons of Condition-based Maintenance  

• CBM can be expensive as you have to consider installation, specific control systems and asset modifications for older equipment.
• There may be a need to invest in staff training as CBM requires a greater understanding of performance criteria, sensors and control systems.
• Maintenance needs to be carried out as and when, as opposed to being predicted in advance. 

Predictive vs Condition-based Maintenance 

 As you’ve found out, there are several elements of predictive and condition-based maintenance that overlap. 

However, there are a few key differences in how data is collected and implemented: 

• Predictive maintenance uses data analytics combined with machine learning or AI to predict failures. 

• Condition-based maintenance focuses on monitoring the current state of an asset using sensors and initiating maintenance when a condition warrants it.

• Predictive maintenance can also identify issues more in advance compared to condition-based strategies. For example, PdM can detect potential problems weeks or months ahead of a failure, while CBM is more likely to be days or weeks before.

Choosing the Right Maintenance Strategy for You 

You now know how and when predictive and condition-based maintenance would be applied, and how exactly they differ.  

At SFG20, we want all FM professionals to have access to straightforward guidance to help them achieve both their responsibilities and building safety with complete confidence.  

If you are uncertain about whether these strategies are suitable for your organisation, you can learn more about other approaches in our other easy-to-understand guides below. 

Expand Your Knowledge