The path to maintenance excellence is a long road that includes a variety of checkpoints along the way. Each organization takes its own route, and some are further along than others. Everyone wants to improve plant maintenance and reliability, but many often don't know where to start, let alone where they want to end.
Wayne Vaughn, former director of maintenance at Harley-Davidson and now a principal consultant at Vesta Partners (www.vestapartners.com), was one of many presenters at Noria Reliable Plant 2012 in Indianapolis this week, and he laid out a route.
Assess, collaborate, build the business case, develop a project plan, execute the plan, and reassess to make adjustments – these six steps are a roadmap to excellence, says Vaughn. “The hardest tasks to do are the strategic ones, the one that are important, but not urgent,” he said. “Those will move you in the right direction.”
Work processes drive costs, explained Vaughn. “If you want to improve the cost of maintenance, you have to change the process,” he said. “Do the right maintenance for effectiveness, and do maintenance right for efficiency. Understand where you want to go.”
The business case must include ROI and must be aligned with the larger organizational needs. “Ultimately, you’re going to have to justify your resource needs,” he explained. In best-in-class maintenance organizations, planned work is 80% or more of total labor, and emergencies account for less than 5% of total labor. Maintenance costs need to be less than 2.5% of the replacement value of assets, he said.
A Holistic Approach and FRACAS
Assessing the condition of machinery is a balancing act. Robert Perez, staff reliability engineer at Enterprise Products and author of “Is My Machine OK?” (http://store.noria.com/Is-My-Machine-OK-A-Field-Guide-to-Assessing-Process-Machinery-P187.aspx), explained how to make the decision on whether to shut down operating process machinery due to a perceived problem or to let it run and risk undesirable consequences. Perez stressed the importance of thinking in terms of the system and not the component.
“What is the sound of one hand clapping?” he asked. “A pump or compressor cannot pump unless it’s installed in a system. You can’t separate the pump from the system. It’s a system issue, and it’s your job to find the weaknesses. Sometimes, the machinery is vibrating, but it’s not necessarily the pump. Usually, it’s a system issue.”
A failure reporting, analysis, and corrective action system (FRACAS) can help to increase machine availability and reduce costs by avoiding failures, or at least managing them, said Jim Taylor, director of operations at Machinery Management Solutions (www.machineryhealthcare.com).
Failure reporting can be done manually, or it can be automated. “Use digital data collection, if you can,” he advised. PLCs, a SCADA system, or a DCS can be used to capture times, parameters, and production information, for example. “Teach maintenance people to write in declarative sentences to convey information or make statements,” said Taylor, who even suggested hiring a local English teacher to train maintenance workers to report failures properly.
Failure analysis should be done to aid prevention, not to ascribe blame, he cautioned. “Use the 5-whys analysis to get to the root cause,” he said.
For the corrective action plan, Taylor suggested using the RCM decision tree. “Make sure the procedure is effective and applicable,” he recommended. “With an effective procedure, it costs less to do the procedure than it costs to accept the failure or consequence. An applicable procedure actually solves the problem.”
The final step is to implement accountability, he added.
Vibration Analysis and Bearing Failure
Everyone knows how important it is to detect fault conditions using vibration analysis, but understanding bearing frequencies and how to approximate time to bearing failure can be tricky. Jason Tranter is the founder and managing director of Mobius Institute (www.mobiusinstitute.com), which maintains a database of 80,000 bearing readings to understand the meaning of a particular frequency, whether it’s cage frequency, ball spin frequency, ball pass outer race frequency, ball pass inner race frequency or others. “Vibration is generated by a bearing when there’s a fault, and it generates a frequency that could mean unbalance, misalignment, or a gearbox problem,” he explained. “But bearing frequencies are not exact.”
It’s important to track how a bearing fault progresses. “There’s a reasonably predictable change in vibration as a fault develops,” said Tranter. “By combining condition-based maintenance with precision maintenance, we can extend the life of the bearing and reduce the risk of unexpected failures.”
Tranter also spoke with Plant Services Senior Technical Editor J. Stanton McGroarty at the end of the day about vibration analysis. See the interview here: http://youtu.be/kVYGlMqkiWA.