Predictive maintenance best practices

Implementing a predictive maintenance strategy is  a learning process that evolves through practical experience. While strategic planning provides direction, success depends on mastering the operational details.

These best practices represent lessons learned from real-world deployments. They help you bridge the gap between well-designed proof-of-concept projects and sustainable, scalable programs:

• Sensor installation and data quality
• Environmental compensation and analysis
• Data translation and team communication
• Maintenance frequency optimization

When it comes to implementing a predictive maintenance strategy, ifm uses an “I do, we do, you do” framework. This means the process starts with your ifm solutions architect evaluating your current needs, assets, and more. They then work with you to develop and launch a PdM project and work alongside you to troubleshoot challenges and address concerns. 

This sets you up to continue the project on your own. Of course, ifm is always available when you have questions, or are ready to expand and scale the program. But, you’ll be confident in running the program without paying for outside consultants. 

With that in mind, this page covers the overall concepts that help a PdM project or program succeed. Consider these basic best practices as your starting framework for building expertise that develops through hands-on application.

Sensor installation and data quality

Sensors monitor equipment performance continuously and return high-quality data when properly installed:

• Sensors require strategic placement: One sensor for equipment under two feet, sensors at both ends for larger equipment. Drill and tap installation provides optimal accuracy, Loctite adhesives work when permanent mounting isn't feasible. Avoid magnetic mounting for continuous monitoring.

• Mounting methodology directly impacts data accuracy and equipment reliability. 

  • Drill and tap installation provides optimal rigid coupling between sensors and monitored equipment. 

  • Loctite adhesives offer the next-best alternative when permanent mounting isn't feasible. 

  • Magnetic mounting serves temporary applications but compromises long-term measurement consistency, making it unsuitable for continuous monitoring programs.

• Avoid mounting sensors on equipment fins, casings, or shrouds. These flexible surfaces amplify sensor housing vibration rather than actual equipment conditions. They create a trampoline effect, leading to false alerts and reduced diagnostic accuracy.

Environmental compensation and analysis

Monitoring systems should calculate delta values between asset readings and ambient conditions rather than using absolute readings. This compensation prevents false alarms during environmental variations while maintaining sensitivity to actual equipment degradation.

Real-time data enables immediate response when properly contextualized for environmental conditions. 

Smart analytics automatically adjust thresholds based on environmental factors, ensuring alerts indicate genuine equipment issues.

For example:

During heat waves, absolute temperature readings in automotive plants in southern climates suggest problems. But, equipment is operating normally relative to high ambient conditions.

Data translation and team communication

Transform technical sensor outputs into actionable language that reflects required maintenance activities. This translation approach eliminates the complexity barrier that can prevent you from successfully implementing predictive maintenance techniques. 

Machine learning models improve over time, but maintenance teams need immediate understanding of current conditions. 

For example:

Our vibration sensors output values like ARMS, VRMS, peak acceleration, and crest factor. These terms don’t resonate at all with most maintenance managers. 

In our system, we rename these parameters to more intuitive descriptions. ARMS becomes "friction" or "lubrication."

When maintenance gets an alert, they don't see confusing technical jargon. Instead, they see that the "lubrication" parameter is elevated, which tells them to send someone out to clean, inspect, and lubricate the equipment.

Team communication also extends to machine operators and maintenance technicians.These staff members work with machines daily. Their working knowledge of each asset and their failure patterns provides valuable information about historical machine conditions and the impact of a new PdM program.

Maintenance frequency optimization

Establish procedures where your maintenance team can address alarm condition alerts and work orders as soon as possible. This is different from predetermined maintenance schedules or run-to-failure scenarios. 

Part of the predictive maintenance strategy is eliminating unnecessary interventions while ensuring critical maintenance occurs before failures develop. Equipment reliability improves when maintenance frequency responds to actual equipment condition rather than arbitrary time intervals. 

How to get started

The best predictive maintenance strategy is simple to implement and maintain. It also prepares a facility for the future. The best way to create a customized preventative maintenence program is to start with a proof-of-concept project. For a free consultation with a Solutions Architect, fill out the form below. 

For immediate assistance, contact our service center: 
info.us@ifm.com
855-785-0325