Through advanced analytics, the oil and gas industry can bring unplanned downtime — a critical variable in profitability — under control.

By Dr. Linda Alrabady

While not a new problem, unplanned downtime is one which can be newly mitigated and prevented. Advanced analytics are lighting up the way for more reliable equipment maintained with financial outcomes in mind.

The Cost and Opportunity of Unplanned Downtime

In a survey of oil and gas companies globally, Accenture found that the industry faces over 23 days of unplanned downtime annually — about 6 percent of total productivity. At $60 per barrel, value lost or deferred because of unplanned downtime events over the course of a year translates to $150 billion in production losses and a 3X increase in maintenance costs.

In recent years, price volatility, calls for decarbonization, growing marketing concentration in the industry, and aging infrastructure have brought about slimmer margins. While operators have seen success undertaking cost-cutting initiatives as a result, they are still unable to stop margins from slipping. 

Even before this last year, many companies posted consecutive years of lower returns on average capital employed. The pandemic revealed the acute need for industrial analytics and remote monitoring for critical equipment. By and large, COVID-19 redoubled existing trends.

Amid the challenges facing the industry, a variable that operators can control in the operating equation is downtime. With equipment performing in predictable ways, oil and gas companies can consistently capitalize on unplanned downtime through cost savings, and in doing so, help address significant challenges industry-wide.

Even a one percent reduction in the oil and gas industry’s capital expenses, operating expenses, and inventory management would result in estimated savings of $140 billion over the next decade according to Goldman Sachs. Targeting unplanned downtime can be a steady way to uncover significant OpEx savings.

To address these challenges, operators first need to get ahead of the preventative and reactive maintenance cycle.

Business-level decision-makers typically have not had real-time insight into plant availability and downtime metrics.
Business-level decision-makers typically have not had real-time insight into plant availability and downtime metrics.

The Preventative and Reactive Maintenance Cycle

The oil and gas industry has tried different maintenance strategies for various equipment types. Much of the approach centers around time- or use-based intervals. A set schedule determines when technicians conduct inspections and perform maintenance. In the event of an unplanned downtime, corrective maintenance is quickly taken to bring back the equipment online. This approach towards maintenance works well when the probability of equipment failure increases over calendar time, run time, or cycle count. But many failures are not attributable to age and are random in nature with constant failure rates — meaning that it could happen anytime and applying time-based maintenance strategies here might have a detrimental impact on the overall equipment reliability and availability.

This preventative approach of mixing routine inspection and repairs as well as reactive service, is often the default operating mode for many oil and gas companies. One downside to this method is the risk of over- or under-maintaining critical equipment. It’s also why business-level executives at many asset-intensive companies have traditionally thought of maintenance and reliability as cost centers.

Two Paths to Mitigate Unplanned Downtime

Among many possibilities, there are two potential paths to better evaluate preventative maintenance strategies. One way is through validating tasks in a reliability-centered maintenance (RCM) study. With industry-proven strategies, companies can improve equipment reliability and uncover maintenance savings even quicker.

Another way, usually done in concert with preventative maintenance strategy validation, is through advanced analytics powered by data. With an emphasis on data, advanced analytics can detect issues with equipment well before they result in extended periods of unplanned downtime or catastrophic failure. Proactive root cause analysis can become a tool for higher-value inspections and maintenance tasks. Or advanced analytics can simply validate PMs in place. Either way, organizations have the data to inform enhanced visibility into maintenance strategies.

More advanced analytics that open up the opportunity for more profitable maintenance, including predictive analytics, enable teams to pre-empt unexpected downtime events and treat pending asset conditions with proactive, scheduled services. Bringing together data reflective of equipment conditions now and historically can allow data scientists to identify dependencies and make learned assumptions about the potential failure signals and usage patterns leading to failures. Through the more precise understanding of equipment conditions as well, the repair can be made efficiently the first time.

By contextualizing equipment data from operational technology (OT) systems with IT information about OEM guidelines and work order history, maintenance and reliability practitioners can have an equipment-specific maintenance approach that better contributes to its productivity and the lifetime return on the asset.

Advanced Decision Support

With advances in cloud computing and enterprise data stores, there are new opportunities to improve equipment maintenance and reliability across the enterprise. That way, oil and gas companies can ensure their plants are available and making as much money as possible through informed planning decisions powered by advanced analytics.

Dr. Linda AlRrabady Uptake
Dr. Linda Alrabady

Dr. Linda Alrabady is a Senior Manager of Reliability at Uptake, an Industrial AI and Analytics company. Before joining Uptake, Linda was a Reliability Engineering Superintendent at BP. She holds a PhD in Reliability Engineering and Artificial Intelligence from Cranfield University.