Somewhere in North America today, a pump that is responsible for delivering residential drinking water through a municipal pipeline will stop working. Whatever the cause, the result will be largely the same: water supplies will be diverted, lives will be disrupted, and emergency repair funds will be spent.
The good news is that events like these may not be typical for much longer.
Thanks to a precipitous drop in monitoring technology costs and powerful new predictive analytics, it is now far easier—and less expensive—than ever to avoid pipeline, pumping and other system failures. And municipal water systems are the tip of the iceberg of beneficiaries. New stand-alone predictive maintenance solutions like these can be integrated into new and existing applications across a wide range of industries.
What a Difference a Sensor Makes
Monitoring of pumping systems and pipelines is not new. Water pressure has long been monitored by the nation’s water utilities. Regulations require midstream oil and gas pipeline companies to invest in leak detection systems.
But traditional electronic monitoring processes like these are not particularly effective at signaling operators when system failure might be imminent. That takes data on other warning signs, such as equipment vibration and temperature.
To obtain that kind of data, operators traditionally have been required to contract third-party specialists who survey conditions along a specific route. The data they collect can later be compared to previous reports to identify signs of deterioration. But because the data is collected only intermittently, it can be difficult, if not impossible, to pinpoint when or how a specific problem is triggered.
Today’s advanced remote monitoring systems can deliver all that, and more. By adding sensors to motors, pumps and other equipment, it is now possible to collect continuous, real-time data that suggests a pump is approaching an unexpected shutdown or experiencing conditions that might negatively impact process performance. Depending on the application, that data can include everything from current and voltage information, to flow and pressure monitoring.
This is just the kind of information a maintenance service company or in-house staff can use to prevent unexpected and costly equipment failure, improve performance and lengthen equipment service life.
Predictive Analytics Add Value
This advanced form of monitoring gains even more value when predictive analytics are added to the equation. Using algorithms, the data collected through monitoring can be modeled to identify patterns that help users not only intervene on imminent breakdowns, but predict future behaviors and events.
Thanks to decades of route-based monitoring, some organizations have already amassed considerable data that can be more fully exploited. Using predictive analytics, experts can use this data to pinpoint the historical causes
of past problems and identify the best next steps for impending ones.
Consider how such a system might be used on an aging large motor driving any kind of equipment. These technologies make it possible to set an alarm that alerts a user if, for example motor fan vibration exceeds an acceptable or preset value.
Today’s systems can do even more. A monitoring system can also be designed to alert the user if the reading starts to change and approach that value. With the knowledge that operating conditions have changed, maintenance crews can take corrective measures to address situations that are trending.
The algorithms that support these alerts have another benefit. The predictive analytics born from crunching enormous amounts of data is not subjective. Vibration experts can use this information to make unbiased judgements on the cause of particular problems, minimizing unnecessary or improperly focused repairs.
From Preventive to Predictive Maintenance
Thanks to real-time asset condition monitoring, the ultimate benefit of these solutions is their ability to move organizations from a preventive maintenance model to a predictive approach.
Instead of following a predetermined schedule for maintenance and repairs, organizations that use these solutions can focus instead on the actual issues that are currently or might soon be affecting system performance. A sensor that picks up on changes in vibration, for example, might lead a maintenance team to maintain a motor bearing in lieu of performing general maintenance on a rigid schedule, producing substantial cost savings in the process.
All this information can be accessed from a browser-enabled computer or mobile device at a fraction of the cost of just a few years ago. Sensors and microchips that once cost more than $500 are now available for $25 or less. The cost to transmit and store data in the cloud has also been slashed by as much as 95 percent opening the door to new ways of doing business.
Even so, an organization is wise to develop a monitoring and analytics strategy before jumping in head-first. Rather than monitor everything, it may make better economic sense to deploy these technologies strategically on mission-critical systems with the greatest value or on those that pose the greatest risk.
It doesn’t matter if the system resides in a refinery, a water treatment plant or a manufacturing facility. It’s also not what the equipment is driving, but the impact that it has on an operation if that equipment must be stopped. System designers must consider not only the loss of output, but also the cost of replacement equipment and staff lost time in the event of failure.
Maintenance Industry Implications
Predictive maintenance capabilities like these are projected to have their greatest long-term impact—and create some of the most valuable business opportunities—in the maintenance, repair and operations industry itself. In many cases, they have the potential to alter these third-party provider business models, beginning with the scope of services they perform and the customer base they can reach.
They encourage service providers to become more creative and broaden service offerings to include data interpretation and preventive maintenance approaches. By providing insights on equipment operation, these solutions make the service providers smarter and more efficient at what they do. Improved cost control can help these companies gain a competitive advantage, while also contributing to higher customer satisfaction.
End users benefit from these solutions in a variety of ways. Most notably, they can help extend equipment life and increase up-time. By avoiding even one unexpected equipment failure, these systems often pay for themselves.
But they may also enable end users to negotiate extended product warranties, presuming they agree to provide the predictive maintenance services that the monitoring solutions prescribe. Because these stand-alone systems can be applied to any manufacturers’ equipment, they also have the potential to help end users make better buying decisions. By comparing the performance of equipment from multiple vendors, users can choose the systems with the longest proven service life.
The Tipping Point
The capabilities of these monitoring and analytics solutions continue to expand. But they have already reached a tipping point where they deliver significant benefits to organizations that adopt them.
Real-time remote condition monitoring gives end users a new smart tool for addressing equipment deterioration early on. When used in combination with data analytics, these systems help define optimal time and scope for maintenance and potentially identify the root cause of problems, making service providers even more valuable to their customers. These solutions fuel continuous product improvement at OEMs, while also empowering end-users to make smarter, performance-based equipment decisions. Ultimately, they lead to lower operational costs and longer equipment service life.
The bottom line: If downtime of key equipment carries a significant cost, the return on investment in a stand-alone remote monitoring and analytics solution is typically more than worth the price.