Turning your data historian into a futurist

0

Under-utilized technology creates a drag on an organization. The ability to get more out of the tools you already use can increase the value of an existing investment, and that value grows as processes become more efficient and decisions are based on firmer foundations.

Consider the facilities engineer at an oil refinery. She might run 12 weeks of historical data out of her data historian to monitor process performance, examine productivity gains or losses, or understand why an equipment failure occurred. It’s an invaluable tool for examining past performance. But what if she could harne460818751lorezss these massive amounts of process control data, gathered from every sensor in the operation and logged in the historian, to enable forward-looking decisions?

Data under-utilization goes on across the oil and gas value chain. Very large quantities of data are stored – at high cost – but they are used for small tasks, usually by a single user in a spreadsheet. It's a missed opportunity to enable more users to surface a range of outcomes from the same data, using an analytics platform.

This under-utilization fails to capture the capability of the data to behave as a powerful corporate asset. When combined with predictive analytics, an operational historian’s extensive time-series database can easily be converted into a font of forward-looking insights, such as equipment failure forecasts. This single task can alter maintenance schedules and inventory processes in ways that are highly beneficial to an organization’s bottom line.

That’s because what the data historian collects is an essential element for building a predictive model – and as the historian continues to collect sensor information in real time, the model can continuously evolve to produce more refined outputs. This rich trove of historical data is typically used to monitor production performance or analyze why a recent equipment failure occurred. But that same data can be used to create a behavioral model that illuminates inefficiencies and anticipates what is going to break down when. The result is a broader view of the status of facility operations that offers previously unattainable advantages.

One such advantage is a reduction in expensive and inconvenient unplanned maintenance incidents. Equipment repair and replacement can be scheduled at the most advantageous time, causing the least interruption to a facility’s processes. This same process also improves root cause analysis, which in turn refines the predictive maintenance process.

This predictive use of a data historian also improves inventory management. The need to have a large supply of every conceivable replacement part sitting in storage – functioning as a sunk cost to the company until it is needed – can be converted into an efficient, just-in-time stocking process married to an orderly maintenance schedule where unpleasant surprises are reduced significantly.

Under-utilization of existing technology investments is a hidden cost in many organizations, one that analytics can help transform into better performance of tasks that actually add dollars to the bottom line. As the oil and gas business struggles to limit costs and increase revenue at a time of drastically reduced commodity prices, the potential savings of harvesting operational historian data for predictive analytics would be a welcome addition to any company’s analytical toolbox.

Download the SAS white paper on expanding data historian value in the Internet of Things or read an article about predictive asset maintenance. 

Share

About Author

Ian Jones

Senior Strategist, Energy Risk Management

Ian Jones is Senior Strategist for SAS' energy commodity risk management practice. Prior to joining SAS in 2009, he served as editor of the industry trade journal The Risk Desk.

Leave A Reply

Back to Top