Predictive maintenance aimed at reducing process downtime

Paul Studebaker is chief editor of Control. He earned a master's degree in metallurgical engineering and gathered 12 years experience in manufacturing before becoming an award-winning writer and editor for publications including Control and Plant Services.On a deepwater oil platform in the Gulf of Mexico, they call it "unplanned deferred production." They planned for 1.4% and ended up with 11.5% – $190 million in unplanned deferred revenue. The typical rate of what most people call "unplanned outages," or simply "downtime," costs Gulf rigs "more than $1.5 billion annually," said  Jamie Stapleton, global product manager, Foxboro Evo, Schneider Electric, at the company's 2015 Global Automation Conference, today in Dallas. That's 160% of their total OpEx, and it's not just a problem on oil rigs.

Savvy facilities folks recognize the well established hierarchy of maintenance practices with reactive (breakdown) maintenance at the bottom, scheduled preventive maintenance (PM) and reliability-centered maintenance (RCM) as rungs, and predictive maintenance (PdM) as the best practice.

When it comes to instrumentation and controls, most plants are somewhere between reactive and preventive maintenance, spending their resources on doing a combination of emergency repairs and tasks on a fixed schedule, whether or not they're needed. RCM uses asset information and histories to improve the effectiveness of PM, and it helps, but it can't cope with random failures and outliers, so downtime often remains unplanned.

"Predictive maintenance is uber-profitable." Jamie Stapleton spoke on instrument asset management at Schneider Electric's 2015 Global Automation Conference today in Dallas.

Schneider Electric's Maintenance Response Center software collects and presents that information, supports mobility, provides decision support and is connected into a work order system. "It collects everything in real time, provides a unified view in a common dashboard, and offers asset health information for the operational life of the asset," said Stapleton. "It gives you the insight to make smarter decisions."

Smart analytics identify impending failures and trends toward quality problems. When it determines a problem is developing, Maintenance Response Center gives users clear and understandable alerts with recommendations for action, in context and with criticality per NAMUR 107. A simple alert is least critical, and ascending levels add urgency. "You can see without reading how serious the situation is," said Stapleton.

The software automatically generates a condition summary for a quick overview of assets that need attention. Their problems and recommended actions are described in the asset vendors' own terms, "not ‘bit 27,'" said Stapleton, with the tag number, area, manufacturer, type and timestamp.

You can explore assets by tag, drill down into asset history, and track assets even when they are removed, reconditioned and placed in another area.

Users can directly enter work orders, and assign work to the appropriate technician. A targeted reporting feature improves shift turnovers by presenting the information the next shift needs to minimize disruption. "It makes the next shift far more productive," Stapleton said.

It's a browser-based application, so users can access it from anywhere and take targeted action through mobile devices if desired. Security features limit access to users only, with configurable limits on what they can do. Stapleton said, "It doesn't allow any changes to the field devices."

The system retains knowledge for re-use, so it can capture the experience of aging workforces and makes it available to new technicians.

The initial release is dedicated to Foxboro Evo. The platform is hardware-agnostic -- we're planning to add Triconex and Modicon, and to integrate it with Avantis and other facilities-level asset management systems.