6 Approaches to Maintenance and Reliability in Oil & Gas
Maintenance and reliability teams have a key role to play in accelerating the transition to low-carbon energy. A portion of the gains from traditional energy operations is being reinvested in alternative energy ventures.
The more gains that can be used towards alternative projects, the faster the transition. It’s just one more reason why cost-effective asset management is so critical. It’s clearing the path towards operational efficiency required to make those investments, along with dedicated initiatives in environmental, social, and corporate governance (ESG).
Maximizing returns on production, especially as barrel prices rally from pandemic lows, is enabling energy companies to more effectively manage carbon-intensive processes today and establish the basis for diversifying their energy portfolio.
Amid rising operational and maintenance costs, traditional data silos, and personnel shortages, more cost-effective maintenance — leading to higher asset and process reliability — is essential.
Rising O&M Costs in Oil & Gas
Over the years, businesses have overhauled maintenance processes to alleviate downtime and improve overall equipment effectiveness.
The average facility in oil and gas loses 32 hours of productivity each month to unplanned downtime, at a cost of $220,000 per hour. That rounds up annually to $84 million at each facility. For energy companies, regular O&M costs have significantly increased with aging equipment.
Oil and gas companies are facing rising O&M costs with aging equipment.
Not all oil and gas companies have borne the costs of unplanned downtime equally, and not all asset types warrant the same treatment.
Here are the six general approaches to maintenance in oil and gas, and their degree of impact on business outcomes.
Reactive Maintenance (or Run to Failure)
With reactive maintenance, the equipment is allowed to run to failure. This may make sense for non-critical assets that have a limited impact on worker safety, enterprise risk, and bear marginal repair and replacement costs.
For critical assets that have a significant impact on process productivity and sustainability, running to failure is scarcely an option for safety, risk, and business continuity.
Planned Preventive Maintenance (PM)
Service determined by preventative maintenance strategies primarily happens on a time- or usage-based interval. This often means that maintenance is scheduled while machines are still working in order to prevent unplanned downtime and maximize the lifespan and productivity of the equipment.
While effective in some instances for certain assets and for organizations whose industrial data may be hard to extract and move to the cloud, there are certain drawbacks to using this approach. It’s not an exact science, you run the risk of over-maintaining or under-maintaining your assets, and it relies on manufacturer guidelines for routine checkups but doesn’t take into account contextual information like weather or market prices.
Many machines fail because of multiple variables at play — design, maintenance history, duty cycles, and age, among other contextual factors. These failure patterns may not be apparent to the most seasoned maintenance and reliability practitioners, or even in an organization with diligent, digital record-keeping in enterprise asset management (EAM) software or a computerized maintenance management system (CMMS).
That challenge with figuring out where to focus your attention, amid all the data the possible decisions that could be made — based on asset risk, value, and criticality, in view of the overall fleet — is one of the reasons why unplanned downtime occurs. Maintenance and reliability teams are being pulled in so many competing directions, When downtime occurs, they can expect to face its many consequences, including:
Lost revenue because of offline equipment
Inefficiency with scrap materials, quality issues, and rework, increasing material and labor costs and resulting in delays in production
Lost hours for non-maintenance
Missed product shipments or service failures, resulting in lower customer satisfaction and danger to brand reputation
Worker safety and environmental incidents
Condition-Based Maintenance (CbM)
In the 1960s, the Federal Aviation Administration conducted a series of investigations into aircraft reliability and found that, contrary to conventional engineering practices at the time, that time of usage was not an accurate measure of asset viability.
Engineering practices were called into question by FAA investigators in the 1960s.
Follow-up research determined the importance of equipment inspections and repairs focused on asset reliability, or reliability-centered maintenance (RCM). With safety at risk on planes due to maintenance, the FAA findings set off a re-evaluation of practices in the profession. RCM questioned traditional engineering wisdom and looked to enhance equipment and component productivity (rather than asset life) based on design expectations of performance.
Where reliability-centered maintenance outlines expectations and priorities for maintenance, condition-based maintenance (CBM) is often the shape it takes in practice. CbM monitors equipment and component behavior, providing an opportunity for maintenance and reliability teams to manage the risk of their equipment and systems.
To this end, many maintenance teams have some degree of rule- or condition-based monitoring analytics. These analytics may also incorporate transactional, sensor, and design data to more accurately capture equipment and component functionality.
Subject matter expertise is integral to the development of maintenance analytics.
This group includes a range of technologies, from simple signal thresholds set by subject matter experts (SMEs) to advanced diagnostic tools that detect and identify abnormal activity across several inputs like temperature, pressure, and vibration.
Despite their differences, these analytics share the same goal. Condition-based analytics focus reliability engineering and operations activity toward heavy machinery with issues. When aggregated at the facility, regional, or company levels, they give the enterprise a way to more effectively manage its risk, productivity, asset procurement, and capital versus operational expenses.
Predictive (and Prescriptive) Maintenance
Predictive maintenance improves upon these existing maintenance approaches, allowing maintenance and reliability teams to proactively manage their risk. It takes many industrial businesses incremental steps to reach a place where their people and processes are not yet in a position to do so. Planned preventative and condition-based maintenance programs are often a good place to start.
Predictive maintenance ensures that asset managers have the right knowledge and tools to keep critical assets running at peak performance, in accordance with operational priorities and market goals. Rather than an intensive change in maintenance, predictive maintenance is a guided approach towards cost-saving enhancement of planned PM activity already in place.
Traditionally thought of as a cost center, data-backed maintenance can assure production.
Using data from various sources like historical maintenance records, sensor data from machines, and weather data to determine when a machine will need to be serviced, operators can make more informed decisions about when a machine will need a repair. Predictive maintenance takes massive amounts of data and through the use of software, translates that data into meaningful insights and data points — helping to avoid data overload and guide engineers toward best practices.
As digitization, IIoT connectivity, and Industrial AI have advanced, industrial analytics have enabled models of machine behavior (and therefore the ability to service those machines) that:
filter out a variety of cases for asset conditions and their root causes, with prescribed steps to correct them and the market value of doing so
control for confounding variables that may contribute to an asset condition
incorporate additional benchmarks like historical performance and data from surrounding assets
Predictive maintenance detects potential failures and converts what could have been unplanned downtime into high-value planned PMs. This advanced visibility enables better production and operational planning. For this reason, predictive maintenance and predictive maintenance analytics are often also referred to as prescriptive — they enable operations to plan operations.
Dynamic Maintenance
Today, predictive analytics and maintenance form the outermost boundary of augmented decision-making happening at scale.
In the future, asset management and operations teams should be able to combine all of the above maintenance analytics and their corresponding strategies — planned preventative, condition-based, descriptive, predictive, and dynamic — to optimize operational and maintenance costs, current productivity, asset lifecycle management, and market demand.
This alignment will be used to plan incremental improvements in operating plans and, in turn, the design of plants and equipment. Some have termed this the autonomous or self-optimizing plant.
This is where heavy industries are headed, but we’re not there yet. And for the majority of oil and gas companies, they’re just getting started on their journey from planned preventive maintenance to condition-based and predictive maintenance. Realistic evaluations of capabilities, and the range of data available for such initiatives, are important steps to begin.
The good news is that oil and gas companies can set up a framework to improve their maintenance and reliability — incorporating insights from their industrial intelligence to make decisions at the asset, process, site, and enterprise levels.
Ready to put your data to work?