Condition-based Maintenance: How Does it Work, How to Set It Up & What Are the Best Solutions

Use our sophisticated software comparison tool to quickly shortlist CMMS Software options that meet your condition-based maintenance requirements

Condition Based Maintenance (CBM) is like using your smart watch.

Using sensors to monitor data can improve your equipment’s health. Reduce equipment maintenance costs, and, at the same time, improve your overall maintenance KPI’s.

The central idea of CBM is to take maintenance actions based on equipment condition before it fails, i.e. You are not waiting until failure and not leaving too much life in the equipment. Avoiding early maintenance or early replacement scenarios.

The challenge of CBM, is knowing how to set it up correctly. Fully understand the equipment condition data, and take corrective action.

What Is Condition Based Maintenance?

Condition-based maintenance (CBM) is a process of real-time monitoring of equipment operating condition to identify just-in-time maintenance points based on condition data.

It is vital to know that condition-based maintenace is undertaking maintenance based on the current state of equipment. To clarify CBM is not based on usage, time, downtime or prediction.

The components of condition-based maintenance include embedded sensors (vibration, oil and lubricant analysis, temperature, current, etc.), a software program (like CMMS Software), and set trigger events.

Each component contributes to the 3-step process of CBM:

1. Sensors collect equipment data.
2. CMMS Software analyses equipment data.
3. Maintenance team take decisions based on failure probability of the equipment.

The primary focus of a condition-based maintenance system is to improve the reliability and availability of critical assets. Avoiding early maintenance, over maintenance or reactive maintenance on expensive equipment.

This is as opposed to a singular preventive maintenance plan, which results in unnecessary maintenance checks and replacing parts that still have an optimal remaining life.

Source: www.swri.org

A data-driven, round-the-clock monitoring approach like CBM helps to reduce both unplanned and planned downtime of equipment. Often reducing expensive repair and breakdown costs by 15 to 30 percent.

CBM can also be used to understand safe operating conditions for equipment. Particularly for equipment intensive operations such as Oil and Gas, Manufacturing, and Aerospace industries.

4 Steps to Implement Condition Based Maintenance

1. Pick sensors based on your equipment.
2. Understand maintenance thresholds or use out-of-the-box algorithms to read equipment data collected by sensors.
3. Configure sensors and CMMS software.
4. Take maintenance actions based on threshold/algorithm parameters.

1. Map out Equipment and Sensor Applications

Select equipment that need to be under your CBM plan. As we have noted before, CBM ideally should only be applied to critical high-value equipment.
After selecting your equipment, map out the sensors you are going to use. Below we have laid out some popular CBM sensors by their application.

2. Identify Failure Modes & Set Trigger Events.

The most difficult part of CBM is setting thresholds and understanding data that trigger maintenance events. There can be false triggers!

Operating condition data and maintenance condition data can look very similar. If you are using multiple-sensors, the data reads can be complex and sometimes conflicting.
The most confident way to set normal operating conditions and trigger conditions is using historical asset data analysis.

Failure modes, misalignments, lubricant issues, electrical flux or bearing failures can all be understood by understanding historical data.

3. Configure work order rules in your CMMS.

Trigger events need to be converted into work orders. Often trigger events need to be prioritised and hence the work orders. False trigger events may create false work orders. So it is a good idea to run past trigger based work order setup with your CMMS software vendor too.

If you rely on external maintenance team for CBM equipment, then you need to factor in travel and scheduling arrangements.

4. Start collecting CBM Data.

Now comes the fun part; executing your CBM program. Initially, you’ll be regularly calibrating the way you collect data and the trigger events you set. But, as time goes on, the data you collect becomes more valuable. From this point, you can begin to increase the speed and accuracy of your maintenance and improve your decision-making.

Our expert’s view on monitoring asset conditions

“You really need to understand what assets you’ve got, what condition they’re in, and how that condition reflects the failure modes and effects criticality analysis … we’re able to deploy some digital tools that predict time to failure, predict consequences of failure, and give them the best chance of maintaining those assets so that they don’t fail.”

Tim Flower, former Chief Intelligent Infrastructure Engineer at Network Rail

Condition Monitoring Techniques & Practical Examples

Below are the six most commonly used condition-monitoring techniques:

• Lubricant/Oil analysis: Used mostly in heavy-duty machinery and industrial equipment to monitor machine wear, lubricant condition, and contamination.
• Vibration analysis: Used to measure machine health through abnormal waveforms caused by part misalignment, motor failure, belt slips, or bearing failures.
• Infrared thermology (IR): IR is used to detect thermal abnormalities through infrared cameras and identify what parts are rising in temperature and when.
• Ultrasonic: Typically used in the NDT of aircraft parts, ultrasonic sensors detect high-frequency sounds.
• Current/Voltage analysis: Electrical output and meter readings are collected in real time through the use of power measurement modules.
• Acoustic/Pressure analysis: Used to detect sudden drops in pressure for equipment reliant on gas, liquid, and air.

Example 1: Lubrication Factor at Maintain Reliability Ltd

With poor lubrication resulting in the failure of a third of all bearings, Maintain Reliability Ltd replaced scheduled-based lubrication methods with an ultrasound condition-based strategy. This measured the decibel output value of a bearing, which indicated if the value was above or below the set value. Engineers then knew when to add more grease to lubricate the bearing at the right point before the performance was affected.

Example 2: NDT Ultrasonic Axle Testing

Through ultrasonic axle testing (UAT), a non-destructive testing method, the Australian Transport Safety Bureau identified a fatigue crack on the axle of two freight tanker wagons that had both derailed. The crack initially went undetected and grew, resulting in an overstress factor of the axle.

Example 3: Alek Zelnins, Maintenance Manager at Contract Pharmaceuticals

Alek spent 12 years at Unilever as Maintenance Supervisor and spoke on the Comparesoft podcast about the overall basis of condition-based maintenance:

“Using [condition-based maintenance] technology, you can put a vibration and temperature sensor and monitor the condition of that, as you run the equipment, which is basically the best way of monitoring the condition. Because when the mechanic goes and the work order says inspect that bearing, you have to lock out the machine. You have to kill the source of power. You don’t want to get [the] mechanic injured.

[The] mechanic will inspect that particular bearing while it’s not running, but it’s not a true representation of the state of the bearing. The bearing needs to run in order to be in a current state, whether it’s starting to deteriorate or not. The technology helps us here by monitoring the performance of this bearing 24/7.”

The Benefits of Condition Based Maintenance

Conditon-based maintenance has long been associated with effective maintenance outcomes. Back in 1998, the American trade magazine, Plant Services, found that 86% of plant production owners used one or more methods of Condition Monitoring (CM).

Today, advancements in data-collecting sensors and IoT technology have enabled condition-monitoring techniques to further excel. Now, a CBM program has several benefits:

• Maintenance is only performed when there are signs of a drop in asset performance
• Fewer disruptions to production
• Less unplanned downtime
• More efficient planned downtime
• Reduced maintenance spending – a 1998 study found businesses that implemented a CM program spent 25% less on maintenance
• Faster diagnosis of problems
• A more knowledgeable workforce
• Reduced mean time between maintenance (MTBM)

It’s also worth noting that, without condition-based maintenance methods, there would be no asset portfolio management, performance-based contracts or big data analysis.

What are the key challenges to using Condition-Based Maintenance?

Key challenges of using condition-based maintenance programme are;

1. Knowing how to choose sensors and set-up CBM.
2. Initial costs of the set-up.
3. Understanding trigger data. Failure data can look similar to ideal operating conditions.
4. Can take a long time to determine trigger events for part failures.
5. Often requires experts to retrofit data-collecting sensor devices.
6. Ongoing resources (time and cost) needed to train staff in data analysis and asset performance knowledge.

CBM Vs Reactive Vs Preventive Vs Predictive Maintenance

Reactive/Corrective Maintenance Vs. Condition-Based Maintenance

Reactive maintenance sits alongside maintenance types such as emergency maintenance, corrective, breakdown, and run-to-failure. It is a strategy that is deployed to only repair parts when they have failed.

Although this approach can be cost-saving in the short term, it can lead to excessive long-term repair costs and unplanned downtime.

Predictive Maintenance Vs. Condition-Based Maintenance

The one glaring similarity between CBM and predictive maintenance is the use of data-collecting tools to identify when it’s time to perform maintenance.

The difference is, condition-based maintenance programs use set thresholds and events to determine when maintenance is due. Whereas predictive maintenance uses advanced statistical methods such as machine learning to predict when service will be required.

Preventive Maintenance Vs. Condition-Based Maintenance

Condition monitoring sits under the umbrella of preventive maintenance (PM) alongside failure finding and risk-based maintenance types. But, while CBM is proactive, it is more attentive to real-time data readouts.

A purely preventive maintenance approach, however, relies on scheduled checks and replacing of parts – even if those parts being replaced still have good optimal life remaining. A good example is a Total Productive Maintenance method.

FAQs

1. What Is Condition Monitoring?

Condition monitoring is the process of analysing the real-time readouts of equipment and noting any changes in performance through devices that measure vibrations, pressure, sound, and lubricant levels. It is the basis of a condition-based maintenance program and the subsequent decision-making process when it comes to replacing and repairing assets.

2. Is CBM a Preventive Maintenance Strategy?

Yes and no. Although condition-based maintenance is a form of proactive/preventive maintenance, it more resides with a predictive maintenance program.

CBM is a more focused approach that recommends maintenance only when necessary so that asset performance is kept in an optimal timeframe. Whereas a purely PM approach is set around scheduled and regular maintenance, meaning equipment and spare parts could be replaced even though they have optimal life remaining.

3. Which Industries Typically Implement Condition-Based Maintenance?

Asset-intensive businesses that rely on their physical and heavy assets for productive output are the ones that will benefit most from implementing CBM. These include:

• Oil and gas industry (oil rigs, production plants, downhole monitors)
• Medical operations
• Industrial plants (manufacturing, logistics)
• Automotive companies
• Aerospace
• Commercial (HVAC operators, white goods)

4. What’s the Difference Between CBM and TBM?

Whereas CBM is performed around real-time data analysis from condition monitoring devices, time-based maintenance (TBM) is triggered by a maintenance calendar schedule. TBM is a planned approach to maintenance that is put in place by analysing historical asset data to schedule specific set dates for planned equipment maintenance.

5. What Is Non-Destructive Testing – Is It a Part of Condition-Based Maintenance?

Non-destructive testing (NDT) is a condition monitoring method used for non-moving parts and equipment such as pipes, metals, and structures. NDT is performed through techniques such as ultrasonic testing and eddy current testing to identify defects, damage, welds, and thickness measurements without causing damage to parts.

6. Who decides which sensors to use to monitor the condition of the equipment?

It depends, this could be an on-site specialist or one brought in to identify the modes of failure to be targeted. These modes of failure align with one or more of the Predictive spheres, Vibration is the most commonly used along with Temperature, it might include Ultrasound, Thermography or Oil quality sensors. Whoever holds the budget or Capex ultimately decides. We have seen some very poorly deployed systems due to the lack of knowledge about where to deploy sensors.

7. What are the typical scenarios where CBM and Predictive Strategies are used?

Where the risks and consequences of failure are most costly, most CBM and PdM started in Aerospace and the Oil & Gas sector due to this. If there is an awareness of Reliability Centred Maintenance (RCM) principles then On-Condition based is part of that process. If there’s lots of automation with few Technicians then using sensors can alert to the few percent of assets that may be signalling an issue eg. Nissan UK Bodyshop has 1,000s of robots, conveyors and drives with a shift of only six technicians to look after them; their system looks at everything without deploying sensors to everything.

8. For someone totally new to CBM, what would be your top 3 tips?

1. Read RCM2 by John Moubray is my number one tip to everyone involved in asset care.
2. Use asset history and RCM to identify Failure Modes on the 20% most Critical Assets in your operation.
3. Study Vibration and Ultrasound basics first, then Thermography and Tribology.