As is the case with any new technology, first-time users of condition-monitoring technology must gain a little experience with it to reap the greatest return. John Pucillo, manager of condition-monitoring products at Milwaukee-based Rockwell Automation, offers the following advice on avoiding three common mistakes:
Improper Collection Frequency. Users must match the frequency of acquiring data to failure modes typically exhibited by the type or class of equipment being monitored. For example, collecting data every 30 days will be ineffective for alerting operators to failures that typically occur within a week.
“It is extremely important to apply the right data-collection devices to each class of machine so you can detect failure in the early stages and allow time for predictive analysis and for scheduling the required maintenance,” says Pucillo.
Failing to Act upon Data. As important as collecting data is to condition monitoring, actually using whatever data you collect is far more important. The problem here is that reliability engineers and maintenance technicians tend to do what needs to be done now, such as repairing broken equipment. A variety of solutions exist, but all involve taking the analysis away from folks who are busy with other tasks. For example, you could install intelligent condition-monitoring technology that can detect and diagnose developing problems that can lead to failures, and then send alarms or alerts to appropriate staff. Another option is to hire an analyst. This person could be a full-time employee, but doesn’t have to be.
Underuse of Technology. “It’s not out of the ordinary for users to invest in technology and use only a small portion of its features and capabilities,” observes Pucillo. “The reasons stem from improper specification, poor planning, poor implementation and lack of trained or adequate resources.” He recommends consulting with vendors to ensure that the right level of technology is specified for the application and not too much. He also urges users to discuss the kinds and numbers of people necessary to support the technology and then to work them and their training into their budgets.
See the story that goes with this sidebar: In Condition Monitoring Earlier is Better
