Condition monitoring (CM) is a predictive maintenance approach that relies on real-time data collection to monitor asset/system health and detect faults and anomalies. Organizations that utilize condition monitoring use the approach to identify potential problems before critical assets fail, minimizing unplanned downtime and maximizing asset lifespan.
Typically, the monitoring process involves continuous data collection from various high-tech sensors and instruments installed on the assets the maintenance department wants to track. The sensors can provide a range of diagnostics, including vibration levels, temperature, pressure and sound, among other parameters.
Once maintenance has the data, they can analyze and interpret it using one (or a few) of the myriad techniques and software tools available. The two most common uses for condition monitoring data are:
- Trend monitoring. Trend monitoring uses continuous measurement and data analysis to identify trends that point to asset deterioration. More specifically, the organization decides which metric is the best indicator of equipment health and uses that metric to assess trends in asset performance, and ultimately to anticipate when the asset will deteriorate past the critical limit. Trend monitoring is frequently used to track engine performance.
- Condition checking. Unlike trend monitoring, condition checking relies on periodic measurement of assets while they’re running. Maintenance then uses the data to assess the current condition of the asset. Checking the quality of a machine’s lubricant using an oil sight glass is an example of condition checking.
Regardless of how you use condition monitoring data, you can program your data analysis tools to generate alerts or notifications when potential issues arise. The alerts will trigger the necessary maintenance team or technician to address the fault.
Condition monitoring techniques are most commonly used to keep rotating equipment (e.g., gearboxes, centrifuges, reciprocating machines, etc.). They help organizations optimize maintenance operations, especially in industries—like manufacturing, power generation, and transportation—where machines and equipment are critical to daily operations.
In these industries, even a small malfunction can cause significant financial losses and drops in productivity. For example, in a manufacturing plant, a faulty machine can lead to production delays, missed deadlines, lapses in regulatory compliance and increased costs. In the transportation industry, a malfunctioning aircraft engine can lead to flight cancellations, lost revenue, and even safety concerns.
Ultimately, condition monitoring can help maintenance teams take a more proactive approach to maintenance, saving companies money and maximizing operational efficiency.
Explore IBM Maximo to learn how IoT data, analytics and AI can help streamline your asset operations.
Subscribe to the IBM newsletter
Implementing a condition monitoring program is a relatively straightforward process involving three primary steps.
The first step for implementing a CM program is to collect as much asset data as possible. This should include historical data (i.e., maintenance history) and any documentation from the manufactuer and/or regulatory agencies.
Sensors are the driving force behind any condition monitoring program, so the first step to implementing a program is installing the sensors that will collect the necessary data. Different assets will require different types of sensors and different approaches to sensor installation, so be sure to account for the needs of all your critical assets.
As soon as you’ve installed all your condition monitoring sensors, they will begin to collect machine health data, like vibration and position, rotor speed, temperature measurements and operating process sensors. This data will allow you to establish baseline measurements for assets and decipher what is and is not normal for a piece of equipment.
Assuming your maintenance department has installed machine condition monitoring software, they will assign maintenance data collectors to continually monitor and analyze sensor data to assess asset health and anticipate potential machine failures.
Your organization and/or maintenance team can employ various techniques and tools to implement a condition monitoring program. Common approaches include:
Electromagnetic monitoring
Electromagnetic monitoring measures field distortions and eddy current changes to locate corrosion, cracks, weaknesses and other faults. The technician applies magnetic fields to the asset’s surface walls and tubing to identify faults in surface materials and features.
Infrared thermography
Infrared thermography is a type of non-destructive testing that uses thermal imaging to detect overheating and other temperature-related issues. It uses thermal imaging cameras to capture the infrared radiation emitted by an object/surface and convert it into a visual image (or thermogram). Organizations primarily use this type of CBM to monitor motors, inspect bearings, and check gas, sludge or liquid levels.
Laser interferometry
Laser interferometry utilizes laser-generated wavelengths of light to measure variations from an asset’s baseline wave displacement. Using an interferometer, the maintenance technician measures interference patterns that indicate defects—like corrosion and cavities—in surface and subsurface materials.
Oil analysis
Oil analysis assesses the properties of the oil (e.g., viscosity, acidity, etc.) in an asset to detect contaminants or wear particles. It typically involves collecting a sample of lubricating oil from the equipment and sending it to a laboratory for analysis. Oil analysis can be useful for monitoring assets like engines, gearboxes and hydraulic systems.
Vibration monitoring
Vibration monitoring (or vibration analysis) uses vibration sensors to measure vibration frequencies in an asset and detect abnormalities that may indicate a problem. Since rotating assets (e.g., motors and pumps), for instance, tend to vibrate more intensely and more loudly as they age, measuring changes in vibration can help identify wear and damage before the asset fails. Vibration data can be used to detect a wide range of problems, including misalignment, imbalance, bearing wear or failure, bent shafts and loose components, among other faults.
Acoustic emissions testing
Acoustic emissions testing is a type of vibration analysis, but it involves the use of much higher-frequency sounds to find shocks and cracks. Sound patterns reveal the presence of unusual noises or vibrations that can indicate a problem or impending breakdown. This technique is particularly useful for detecting faults in rotating equipment such as motors, pumps and fans.
Ultrasonic analysis
Ultrasonic analysis (also called ultrasonic testing) uses high-frequency sound waves to detect leaks, cracks or defects in a piece of equipment. It relies on both contact (structure-borne) and non-contact (airborne) data collection techniques to determine asset attrition. Contact methods are typically used to detect mechanical issues—like lubrication issues, gear damage and broken rotor bars—that generate high-frequency sounds. Non-contact methods can detect issues, like pressure and vacuum leaks in compressed gas systems, that tend to generate low-frequency sounds.
Motor circuit analysis (MCA)
Motor circuit analysis, also known as motor testing, uses voltage- or current-based assessments to find electrical imbalances and measure insulation degradation, both of which can cause motor failure. MCA is used primarily to monitor electric motors.
Other methods like visual inspection and performance testing can also be used for condition monitoring. Naturally, each technique has its strengths and weaknesses, so the best choice for your department will depend on your resources, equipment, environment and organizational needs.
Condition monitoring and the IIoT are two closely related concepts that, used together, can improve the efficiency and reliability of maintenance management systems.
Condition monitoring relies on continuous data collection from sensors and other sources to prevent or mitigate problems. The Industrial Internet of Things, on the other hand, is a network of interconnected devices and equipment sensors that communicate with each other and with the cloud to collect and share data.
Using condition monitoring and IIoT in tandem allows for more comprehensive and accurate monitoring, and more efficient communication about maintenance tasks and issues. It not only allows internet-connected smart assets to communicate and share diagnostic data, enabling instantaneous system and asset comparisons, but it also helps teams make more informed decisions about the entire production operation. Furthermore, IIoT makes it possible to collect and transmit data and monitor systems remotely, which can be particularly useful for systems located in remote or hazardous locations.
These features provide maintenance departments more sophisticated analyses, allow them utilize data from multiple machines simultaneously, and helps them automate processes that would ordinarily require maintenance technicians (and their associated costs). Ultimately, condition monitoring systems and the IIoT make it possible to transform the way organizations maintain and monitor key assets, processes and systems, improving the reliability, efficiency and safety of its maintenance operation.
One of the key benefits of condition monitoring is that it enables maintenance teams to implement preventative maintenance management and machine health monitoring. By identifying potential problems before they cause equipment failure, maintenance teams can schedule maintenance activities at the most convenient time, reducing the impact on production and minimizing downtime from unexpected shutdowns.
Condition monitoring offers several other advantages over traditional maintenance approaches, including:
Condition monitoring provides real-time data on the performance of a system or component, which can be used to optimize maintenance planning and scheduling. This helps to reduce the frequency of maintenance activities, while ensuring that they are executed only needed, based on actual system performance.
By detecting and addressing problems preemptively, condition monitoring helps to extend the lifespan of equipment and components, reducing the need for costly replacements or repairs, and maximizing the ROI of assets.
Condition monitoring can help to identify inefficiencies in a system or component, such as excessive energy consumption or unnecessary wear and tear. By addressing these issues, operational efficiency can be improved, leading to reduced costs and improved productivity.
Condition monitoring can help to identify potential safety hazards, such as worn or damaged components, before they cause harm to personnel or equipment. This helps to improve overall safety and reduce the risk of accidents and injuries.
While condition monitoring can really help an organization streamline their maintenance management systems (especially in the case of IoT-enabled condition monitoring), it does have disadvantages that organizations should consider, including:
Implementing a condition monitoring program can be quite expensive, as it typically requires the installation of sensors and other monitoring equipment, as well as an investment in data analysis software and personnel to manage the program and equipment. The cost of executing a condition monitoring program may be prohibitive for some organizations, particularly smaller ones.
Condition monitoring can be complex, requiring specialized knowledge and expertise to set up and manage. Some organizations may not have enough trained personnel to run the system effectively and may therefore need to hire specialized personnel or outside consultants. Furthermore, condition monitoring systems rely on high-tech sensors to run diagnostics on organizational assets. Older facilities without adequate infrastructure may require extensive retrofitting.
Condition monitoring systems can generate a large amount of data, which can be overwhelming to manage and analyze. It can be difficult for maintenance teams to triage data and identify the most important data points and trends.
Intelligent asset management, monitoring, predictive maintenance and reliability in a single platform.
Enhance your application performance monitoring to provide the context you need to resolve incidents faster.
The IBM blog breaks down the difference between condition monitoring and predictive maintenance.
Read how IBM Research is furthering the development of predictive maintenance in technology and banking
More than two-thirds of data that organizations collect goes unused resulting in performance and reliability issues.