Predictive maintenance is maintenance that directly monitors the condition and performance of equipment during normal operation to reduce the likelihood of failures. It attempts to keep costs low by reducing the frequency of maintenance tasks, reducing unplanned breakdowns and eliminating unnecessary preventive maintenance.
With predictive maintenance, organizations consistently monitor and test conditions such as lubrication and corrosion. Methods for accomplishing predictive maintenance include infrared testing, acoustic (partial discharge and airborne ultrasonic), vibration analysis, sound level measurements and oil analysis. Computerized maintenance management systems (CMMS), condition monitoring, data integration, and integrated tools and sensors can also facilitate success with condition monitoring.
For example, CMMS empowers companies to define boundaries for acceptable equipment operation, import readings, graph results and automatically trigger an email or generate a work order when boundaries are exceeded.
Predictive vs. preventive maintenance
Though the best maintenance programs include a balance of both, preventive maintenance and predictive maintenance are different strategies. Preventive maintenance is determined using the average or expected life cycle of an asset, whereas predictive maintenance is identified based on the condition of equipment.
While predictive maintenance is more complex to establish than a preventive maintenance schedule based on manufacturer recommendations, it can be more effective for a business to save time and money. For example, taking vibration measurements on an electric engine at recommended intervals more accurately detects bearing wear and allows organizations to take action such as replacing a bearing before total failure occurs.
How does predictive maintenance work?
Predictive maintenance evaluates the condition of equipment by performing periodic or continuous (online) equipment condition monitoring. Most predictive maintenance is performed while equipment is operating normally to minimize disruption of everyday operations. This maintenance strategy leverages the principles of statistical process control to determine when maintenance tasks will be needed in the future.
For example, rather than changing a vehicle’s oil because drives hit 3,000 miles, predictive maintenance empowers organizations to collect oil sample data and change the oil based on the results of asset wear. For predictive maintenance to be effective, it requires both hardware to monitor the equipment and software to generate the corrective work order when a potential problem is detected. Specific types of predictive maintenance include:
- Vibration analysis: Vibration sensors can be used to detect degradation in performance for equipment such pumps and motors.
- Infrared: Infrared cameras are often used to identify unusually high temperature conditions.
- Acoustic analysis: Acoustic analysis is performed with sonic or ultrasonic tests to find gas or liquid leaks.
- Oil analysis: Oil analysis determines asset wear by measuring an asset’s number and size of particles.
Additionally, tools such as CMMS, condition monitoring, connected tools and sensors, and data integration can help companies act on the analytics collected by these devices and sensors.
Predictive maintenance tools integrated into a CMMS
Whether you need to track assets through oil viscosity, temperature or vibration, the tools within CMMS systems can help develop accurate predictions when a piece of equipment will require maintenance or replacement.
- Condition Monitoring: Within CMMS systems, condition monitoring tools help empower organizations to execute on predictive maintenance programs. Users can define boundaries of acceptable operation for assets and auto-generate work order or emails when readings fall outside of predefined boundaries.
- Connected sensors & tools: These can offer real-time data streams to track events from anywhere and view AC/DC voltage, current, power and temperature data. By wirelessly syncing measurements taken using handheld tools and comparing them to condition monitoring data, organizations can gain the full picture of equipment efficiency and health.
- Data integration: Data can be integrated into CMMS functionality to enable the completion of seamless workflows on a mobile device. This allows maintenance teams to respond to fault notifications while they are on the move, and then they can create, access or process work orders related to the notification in real time. Planned and unplanned maintenance is better coordinated, unscheduled downtime is reduced and response times to problems or systems failure are improved.
How much can you benefit from predictive maintenance?
Studies have shown that organizations spend approximately 80% of their time reacting to issues rather than proactively preventing them. Predictive maintenance puts predictive maintenance ahead of the game. It helps predict failures and actively monitor performance. As a result, it saves time and money. Organizations that commit to a predictive maintenance program can expect to see significant improvements in asset reliability and a boost in cost efficiency, such as:
- 10x Return on Investment (ROI)
- 25-30% reduction in maintenance costs
- 70-75% elimination of breakdowns
- 35-45% reduction in downtime
- 20-25% increase in production
The best predictive maintenance programs take time to develop, implement and perfect. The timeline to achieve gains such as these varies, but some clients see positive returns in as little as a year.
Advantages & disadvantages of predictive maintenance
Predictive maintenance requires more time and effort to develop then a preventive maintenance schedule. To be truly effective, employees must be trained on how to use the equipment and interpret the analytics they pull. However, once the commitment is made, predictive maintenance can revitalize not only a maintenance team, but an organization as a whole. There are condition monitoring contractors who can perform the labor required and analyze the results for your organization.
What to do when predictive maintenance does not make sense
Sometimes predictive maintenance is not the answer to maintenance woes. It might not be the most cost-effective method to manage all assets with predictive maintenance. For example, changing light bulbs on the plant floor. Rather than running diagnostics on the bulb, leveraging a run-to-failure strategy (waiting until the light bulb goes out to change it) makes more sense. There are a few factors to consider when identifying which assets should be considered for predictive maintenance:
- What is the impact on production if the asset failures unexpectedly?
- Can cost-effective tasks be performed proactively to prevent, or to diminish to a satisfactory degree, the consequences of the failure?
- What is the average cost of repairing this asset?
Applications of predictive maintenance
There are many applications of predictive maintenance in a wide variety of industries such as:
- Finding three-phase power imbalances from harmonic distortion, overloads, or degradation or failure of one or more phases
- Identifying motor amperage spikes or overheating from bad bearings or insulation breakdowns
- Locating potential overloads in electrical panels
- Measuring supply side and demand side power at a common coupling point to monitor power consumption
- Capturing increased temperatures within electrical panels to prevent component failures
- Detecting a drop-in temperature in a steam pipeline that could indicate a pressure leak.
How to implement a predictive maintenance strategy
Implementing a predictive maintenance program should be a methodical process from start to finish. The key is to have a long-term view of what to do in order to put all of the foundational components into place.
- Design the predictive maintenance program: Get positive buy in from management and be prepared to discuss and quantify the benefits and goals. Identify which equipment to target for the program by taking a close look at equipment failure histories and the associated root causes. Equipment that is failing the most will provide the most potential for cost reductions and reliability improvements. Compare the cost of implementing a predictive maintenance to the average cost of equipment failures. As stated above, sometimes predictive maintenance does not make sense. Depending on the asset, a corrective method of maintenance could be cheaper.
- Select predictive maintenance technology: Choose which of the above technologies would be most effective to monitor the condition of your equipment. Is your organization more interested in vibration analysis, infrared thermography, ultrasonic inspection or oil analysis? Select the tools that will provide that information.
- Allocate proper resources: Develop and train an implementation team to perform predictive maintenance activities. Carve out time in the schedule for predictive maintenance tasks such as data collection, analysis, reporting and tracking, and allocate funding for predictive maintenance technology investments or, for a predictive maintenance, contractor to assist.
- Perform system integrations: Leverage the tools within and integrated into a CMMS to help turn condition monitoring data into action. For example, a company offering equipment monitoring services, lubrication engineering and reliability engineering can record negative diagnostic reports and automatically generate corrective work orders.
- Coordinate preventive maintenance & predictive maintenance programs: Leveraging both preventive and predictive maintenance makes for the best maintenance programs. Use each method where applicable and decide which strategy to apply based on disruption due to equipment downtime, cost of parts and labor time, and equipment history.
- Utilize CMMS reports & dashboards: With reporting and dashboard tools, organizations can consistently document work order history, failures, costs and trends. This helps to track progress for key stakeholders.