Summary

Engine failure? Burst pipe? High emissions? It's time to alert the service engineers for repairs, right?

Wrong. The best time to service industrial assets is before they fail.

Get ready for sensor-fitted self-monitoring machines that can talk to each other and schedule preventive maintenance checks remotely. That’s the promise of predictive maintenance.

Machines that heal themselves

Take gas turbines, for instance. They are getting fitted with sensors that record and transmit real-time data—and its context—to cloud platforms. Knowledge management prognostics and analytics tools will use artificial intelligence and machine learning algorithms to analyze this data and schedule the visits of field service engineers in a timely manner.

Armed with smart wearable tools rendering digital workspace technicians and their robot buddies will collaborate to complete the required maintenance tasks at the optimum time before issues become disruptions. Always-connected smart machines will continue to share data about maintenance tasks and their results, thanks to the Industrial Internet.

This self-healing equipment might sound futuristic, but it will be a reality soon. More and more machines, equipped with sophisticated sensors, are already plugging into the Industrial Internet. This vast network of smart machines and analytics algorithms will take maintenance way beyond repairs and solve problems before they occur.

Reduced downtime

Turbines are not an exception. As was discussed at the recent Field Service USA 2015 conference, collaboration between smart machines, service personnel, and analytics driven insights facilitates anticipating faults and repairing them before faulty parts affect the performance of the machine.

Downtime can be expensive for any industry. The transportation sector loses $400 million per year due to mainline failures and excess crews. Delays and cancelations cost airlines $45 million per day. The oil and gas industry loses $800 per day for refinery processing. Cutting downtime with predictive maintenance is not just a matter of convenience; it has a real impact on profitability.

Flexible on-demand workforce

This efficiency and predictability will mean a huge shift in field service roles. To begin with, mobile devices and advanced analytical platforms will become as important to field service engineers as their toolboxes. In next two years, smart sensors, analytics-aware processes, and behavior analytics-driven workflows will pave the way for built-in safety and security. In addition the deployment of automated knowledge capture and sharing tools will lay the foundation for change for the field services in general. And in the longer term, field services could even evolve into a crowd-sourced on-demand pay-per-use service model that will tap into a global, mobile, and flexible workforce. This could radically change the lives of hundreds of thousands of technicians around the world while simplifying the end user experience and unlocking new values for customers.

Preventive maintenance for profit

As companies accumulate more data from connected products and field service calls, maintenance will gradually move beyond repairs. Predictive analytics data will get shared with product development teams and research and development centers. These actionable insights, from multiple machines integrated by predictive data analytics in the cloud, could help make better decisions to optimize operations and squeeze better performance from new models of existing products.

With a network of connected devices assimilating into the Industrial Internet, the stage is set to elevate the maintenance of industrial equipment from a cost center to a profit center.

About the author

Piyush Modi

Collaboration and Mobile Lab Manager at GE Software

Related insights