Predictive maintenance 101

It’s no surprise that plant maintenance can be time-consuming and expensive. It also is understood that proper maintenance practices are critical to help maximize equipment uptime and protect the bottom line. However, many practices take the 'break-fix’ approach, including preventive maintenance, performed on a time-based or scheduled scenario regardless of actual need.

By Mark Roether, The Timken Company March 15, 2008

It’s no surprise that plant maintenance can be time-consuming and expensive. It also is understood that proper maintenance practices are critical to help maximize equipment uptime and protect the bottom line. However, many practices take the ‘break-fix’ approach, including preventive maintenance, performed on a time-based or scheduled scenario regardless of actual need. With advancements in technology and services to improve a plant’s productivity and machine uptime, a new maintenance method is gaining ground.

Defining predictive maintenance

Predictive maintenance (PdM) is a method for testing machines that identifies faults at an early stage in their development. This testing is performed without the need to take the machine offline for inspection as is typically required in preventive maintenance programs. Unique in its approach, a quality PdM program utilizes proven testing techniques such as vibration analysis, infrared thermography, oil and wear particle analysis, ultrasonic testing, video imaging, balancing and alignment and electric motor testing.

A well implemented PdM program is designed to eliminate the guesswork from maintenance. With testing equipment that can clearly identify both the problem and potential cause, trained technicians and analysts can recommend best practices and actions to help eliminate repetitive problems, prevent unscheduled downtime, extend machine life and improve overall performance in plant operations and equipment.

The goal of this improvement in machine reliability is an increase in overall plant capacity without the need for huge capital expenditures to expand production capacity. In today’s global economy, this can be critical in maintaining a company’s competitive edge.

Challenges of PdM programs

The PdM theory is easy to understand, but that’s not to say it doesn’t pose challenges. One challenge is going beyond theory to understanding the tangible side of what it means and what it can reveal about the operating equipment.

The most successful PdM programs have the support of upper management. However, put together properly, and with appropriately set and understood expectations, the value and ROI often far outweigh the initial budgetary concerns.

If PdM programs fail, it is often due to lack of upper management support. The right PdM supplier can help answer questions, address budgetary concerns and document tangible results that help show the value of such a program to the bottom line.

Various other reasons PdM programs may not be successful include:

Lack of interaction between PdM technician and plant maintenance group

Downturn in business and additional budget restraints

Lack of tangible results (especially compared to expectations)

Failure to document results (lack of awareness of ROI).

Another challenge is that which faces the company that has tried to implement a PdM program previously but was displeased with the results. It is important to note that not all PdM programs are created equal. Suppliers can vary greatly in experience, technology, integration techniques and reporting. Knowing this, it is important to carefully evaluate the needs of the plant and align those with the proper PdM services that will add value to plant operations.

Implementing PdM

The needs of each plant vary so widely that each program is customized. Now more than ever, companies are packaging their expertise and services to address the market needs and then work to supply, integrate and provide overall reports to customers as defined to meet the goals of the program.

The first step is having a qualified person come to the plant for a walk through to review the equipment and needs and make recommendations for a PdM program. If a PdM program has been in place in the past, it is important to share how it was done and why it was ended. Once the recommendations are in place, budgetary issues and expectations can be addressed.

Carefully selecting a quality program collaborator or supplier can be the difference in the success of the program. A supplier should be judged on quality of service, including the training and expertise of the technical personnel they can provide to staff the program. The field personnel should not only be well trained to collect data but also be trained as an analyst to be able to troubleshoot on the spot.

Three common scenarios for programs are fully outsourced, in-house integration, and a combination of the two.

Fully outsourced PdM programs are often chosen to get the most immediate results and address the concerns of training, technology and manpower. When choosing to outsource, it is recommended that all the PdM services and analysis are integrated with one supplier, so a combined course of action to maximize the effectiveness of the program can be realized.

Some plants opt to have more control over their programs by purchasing the equipment and having in-house personnel trained to manage the program, data collection and analysis. Dedicated personnel, consistency and proper training are key elements to this type of PdM program. Many programs fail initially due to the extensive training required to become proficient at PdM. In some cases, that training can take two years. During that time, results can be erratic at best. This has led to the advent of a third approach to starting a PdM program.

The third common scenario is to outsource initially to get immediate results and then transition to an in-house program through training and support of the initial PdM supplier. This approach is gaining in popularity and tends to yield the best short- and long-term results when implementing an in-house program.

Maximizing the value and ROI

Once the program start up approach is determined, it is important to know how to maximize and demonstrate the value achieved. The first step is to understand exactly what the program brings to the table. An obstacle for many plants considering, or those who have tried PdM, is the vast amount of information and knowledge required to fully understand and gather value from the program.

More than identifying a damaged bearing, PdM equipment can do much more, including:

Find defective bearings long before defects are visually seen

Find misalignment between two rotating pieces of equipment

Recognize when fans become unbalanced

Identify when bearings need lubrication

Tell when an electrical connection needs to be tightened

Alert when oil is contaminated or in need of replacement.

To better explain the services available through PdM, following are examples of the technology:

Vibration analysis — This has become one of the most powerful tools in the industry today to evaluate rotating equipment and quickly determine the condition of the machine’s bearings, alignment, balance and general “health.” Data collected by either continuous or portable vibration equipment is then analyzed to identify the specific machinery fault.

Infrared thermography — Captures variations in heat on both electrical systems and rotating machinery. Non-intrusive, this technique transforms surface temperature information into visual pictures for analysis. This can help track patterns and unusual temperature levels that may require immediate maintenance of electrical connections or help identify “hot spots” in furnace refractories.

Oil and wear particle analysis — Lubricating fluids often can reveal the type and amount of wear on internal machine components. Particles can indicate contamination by dirt, coolant, moisture or improper equipment usage or overloading. All these indicators can help identify potential failures within a machine.

Ultrasonic testing — This testing can help detect air leaks in hoses, pipes, valves and other components before they would be discovered through visual inspection. Tests are conducted during operation to provide more accurate estimates of leakage. This technology also can be used to supplement vibration and infrared thermography.

Video imaging — Used to see inside equipment such as hard-to-dismantle gear boxes, video imaging can detect problems with equipment. Often conducted during routine maintenance shut downs, components such as gears, bearings and other critical parts can be inspected without disassembling equipment.

Balancing and alignment — When industrial rotating equipment is out of alignment or balance, other machinery components may become damaged. Noise is often an indicator of these issues, but proper predictive maintenance personnel can help with detection before severe damage can occur and offer consultation on how to realign components, reset the balance and deter catastrophic failure.

Electric motor testing — Electrical defects are often the root cause of motor failures. This testing can be performed with the motor in place. It quickly provides more information regarding the health of an electric motor. It can be used for quality assurance upon receipt of a repaired electric motor.

Combining these technologies with expertise and product support, many plants and customers are seeing incredible returns on PdM programs.

PdM programs were implemented at four Timken steel plants in Canton, OH in 1997. They now routinely see a 300-600% return on investment. The program is fully outsourced to Timken’s Reliability Services team and is managed by three full-time Timken Reliability Services personnel on site at the steel plants.

Author Information

Mark Roether is reliability solutions sales manager for The Timken Company. He can be contacted at