- Identifying trends to improve your PdM programs effectiveness
- How to determine and track failure modes by PdM technologies
- Performing pareto analysis of the trended data
- Identifying areas to add precision maintenance techniques
- Reducing predictable problems through analysis
The Value of Predictive MaintenanceIf a manufacturing or processing business is to be successful, the reliability of the equipment it depends upon must not be taken for granted. A failure of just one machine integral to a production process can result in significant losses in terms of repair costs, downtime and throughput, and thus it is crucial for operators to take steps to foresee breakdowns or inefficiencies whenever possible. For this reason, many companies have turned to predictive maintenance (PdM) partners to detect potential issues and determine solutions before emergency situations ever occur, allowing for more effective planning. Predictive Maintenance programs use special measuring devices to analyze the health and functionality of every component of a specific piece of critical machinery. If functioning properly the program provides early warning of pending failure and providing long-term assessment of equipment condition to help plan for maintenance and avoid downtime during normal production hours. When we analyze a piece of equipment, we utilize methods such as vibration analysis, infrared inspection, oil analysis and sensory inspections to determine what risk factors exist for that specific unit. For example, we may find flaws are developing in a rolling element bearing that will eventually fail, and will notify the equipment owner to allow for repairs or replacement to be scheduled conveniently, not during an emergency. Beyond offering savings by greatly reducing the likelihood of in-service equipment failure, PdM is helpful in avoiding unnecessary costs related to premature maintenance on functional machinery. By determining when maintenance is truly needed based on the condition of equipment instead of adhering to a generic, calendar-based maintenance program, machine downtime is avoided and service costs are minimized. Additionally, servicing a piece of equipment only when necessary reduces the likelihood of issues occurring as a result of human error during service-related interactions with machinery. Predictive maintenance is of definite value for helping companies avoid costly problems with their equipment, but it often results in a temporary solution to a problem. The PdM process can be cyclical if potential issues are only identified and fixed for the moment. Sometimes the root cause of equipment problem is not addressed as part of the PdM process, resulting in future of redundant issues. While a substantial amount of money is still being saved through the PdM engagements when compared to an instance where equipment failure actually occurs, there is still opportunity for more extensive savings if root causes of problems are investigated and long-term solutions are implemented based on well-supported findings.
The Next Step: Increasing ReliabilityIt is here that the idea of reliability analysis and subsequent consulting comes into play. Every time a PdM group analyzes equipment, they have an opportunity to record what risk factors were detected, how often, and how expediently problems must be addressed based on potential severity. By tracking historical data for every piece of equipment analyzed, it becomes possible to determine trends in the equipment’s behavior and devise strategize to address recurring problems as a result.
Figure One – Percent problems found benchmark for equipment classesFor instance, if a belt drive used to turn a part within a machine is a cause of increased vibration levels, a predictive service program using vibration analysis techniques may be able to isolate the problem and determine the recommended corrective action. If this finding is recorded reliably within a database (see figure two), it becomes possible to look back at the history of the machine and determine whether there are repeat issues with belts, or whether it is simply one of a variety of issues experience over time with that unit. In the event that it is determined that there is a history of belt issues across all machines, an opportunity presents itself to go beyond simply suggesting replacement of the belt—but can choose to also ask the obvious question: “Why?”
Figure Two – Vibration program failure mode review over three yearsThe first step in taking your PdM program to the next level of reliability is changing the mission associated with the program. The idea of “repair before it breaks” need to be replaced with “Don’t repair it, eliminate the root cause”. The basic parts of the PdM program functionally work the same, the routine data gathering, analysis and subsequent reporting. However, enhancing how the reported findings and corresponding failure modes are tracked and periodically analyzed to identify recurring issues or problem trends. As these trends are established, a deeper dive into the root causes would then be conducted and a recommended course of action developed. Many of these issues, based on experience, have been found to be a lack of training and/or precision in the maintenance practiced. It is at this point that the PdM group moves beyond simply providing PdM services to become a true reliability partner. No longer concerned with dealing in the “here and now” and only identifying potential issues, a reliability partner looks to the future by consulting recorded data and positing why failures or potential failures are happening. The service provider can then collaborate with a customer to determine what steps can be taken to avoid the need for maintenance and parts replacement in the future. Consider the case of the belt drive. If it is discovered that a PdM provider has indicated several times in the past that belts are being worn out quickly, they can take their research a step further, asking, “Is this belt lasting as long as it should be according to the manufacturer? If not, is another factor causing failure?” It is frequently the case that faulty parts are not responsible for creating risks: A human element is often to blame. In the case of the belt drive example, belts generally require proper tension and alignment be applied to be effective—something easily thrown off by improper installation, training or tools, not poor equipment design. If this is found likely to be the case, steps could be taken to alter procedures or retrain employees on proper techniques or ways to add precision, ultimately diminishing the need to replace the belts and, in turn, diminishing the amount of money lost in relation to the purchase of new parts and machinery downtime. In the end, the unit’s reliability is improved long-term.