Unplanned shop floor downtime is an expensive fact of life for any manufacturer. In the oil and gas industry, though, downtime is not just costly, it's prohibitive. Oft-quoted estimates put the hourly cost of unplanned refinery downtime at $50,000 or higher.

With that kind of money at stake, oil and gas companies naturally are keen on using software as a weapon in the never-ending fight to reduce or eliminate unexpected downtime. But, where once it was enough to use enterprise asset management (EAM) applications to assign work orders and track assets, advanced oil and gas companies today are shifting their focus to manufacturing reliability, an umbrella term for an effort to maximize plant uptime.

Asset Self-Awareness

A big part of manufacturing reliability is predictive maintenance. First practiced in industries such as oil and gas, where downtime costs are crushing, predictive maintenance involves monitoring the physical health of an asset (wirelessly or not) and sending alerts — in some cases machine to machine — indicating that attention is needed before a breakdown occurs. (For more on machine-to-machine technology, see this month's Special Report, M2M Speeds Up.) The genius of predictive maintenance is that it allows companies to avoid time-consuming, expensive preventive maintenance, which is based on historical data on failure rates but does not take into account the current status of the asset.

Many manufacturers begin to take a more proactive approach to maintenance by implementing scheduled maintenance programs, which are then followed by preventive maintenance programs. In predictive maintenance, the next step, the asset automatically alerts personnel or an advanced enterprise asset management system when its condition is degrading and needs attention. This not only lets you avoid a breakdown, it lets you skip unnecessary, time-consuming scheduled and preventive maintenance.

That's especially valuable in the oil and gas industry.

"Oil and gas used to be very cyclical, with ups and downs in production," says Alison Smith, senior research analyst at AMR Research Inc. But with demand increasing even as prices for oil and gas sporadically rise, she says, these companies are basically at capacity. "They can't absorb a lack of availability of assets. There is not a lot of slack. They have to be up and running continuously." Smith says the cost of downtime in the oil and gas industry runs to millions of dollars per day on average. Downtime can also have safety and compliance implications.

To make the transition from enterprise asset management (EAM) to manufacturing reliability, oil and gas companies are using a host of tools, including traditional EAM packages from vendors such as Indus and MRO Software Inc. (an IBM company), process automation platforms from providers such as Emerson Process Management, ABB, and Invensys plc, and add-in condition monitoring tools from software vendors like SmartSignal Corp. and Ivara Corp.

Many of these vendors are beginning to support predictive or condition-based maintenance, the most advanced approach to maintenance a manufacturer can take.

SmartSignal's EPI*Center software, for example, monitors every aspect of the asset from a reliability, availability, efficiency, and compliance perspective, according to company officials. EPI*Center uses technology that SmartSignal calls similarity-based modeling (SBM) to analyze historical data and construct empirical models of normal equipment and process operations. During real-time asset operation, EPI*Center compares the "expected" sensor values from SBM to real-time data collected from the equipment. By analyzing data from all correlated sensors for a piece of equipment or a system, the SBM technology identifies subtle process deviations well within the normal threshold alarm limit, according to the company. This brings difficult-to-identify problems to the attention of those who can fix them, long before downtime occurs.