May 2018


Digital: Maximize return on capital investment with predictive maintenance

Oil and gas professionals are well aware of the pressures companies face in the market.

Newton, M., AVEVA

Oil and gas professionals are well aware of the pressures companies face in the market. Erratic commodity prices make it extremely difficult to maintain profitability. Environmental regulations have increased the cost and difficulty of regulatory compliance. An aging workforce makes it more difficult for companies to maintain a high level of operational efficiency as aging workers retire and are replaced by their inexperienced counterparts.

In this environment, maximizing asset reliability and performance is a competitive necessity. Powered by the Industrial Internet of Things (IIoT), companies can digitally transform how they manage their assets. Predictive maintenance solutions take advantage of machine learning and advanced pattern recognition to detect anomalies days, weeks or even months before a traditional alarm might provide notification, ensuring that appropriate corrective action can be taken before an incident occurs.

A leading producer of industrial gases saved more than $500,000 from a single predictive maintenance catch. By implementing this technology, oil and gas companies can improve asset reliability, increase operations productivity, reduce maintenance costs and generate a maximum return on capital investment.

FIG. 1. The maintenance maturity pyramid.

The maintenance maturity pyramid. It is important to understand where predictive maintenance fits within a best-in-class maintenance strategy. The maintenance maturity pyramid (FIG. 1) is a useful visual indicator of the different types of maintenance. All strategies across the maintenance maturity pyramid play a part in improving asset performance and reliability.

The base of the pyramid is reactive maintenance, which is simply a run-to-failure model: an asset is run until it malfunctions, at which point it must be repaired or replaced. While reactive maintenance might appear to be the easiest and cheapest maintenance strategy, it is suitable only for non-critical assets.

The next level is preventive maintenance, which is scheduled maintenance conducted at regular intervals. While preventive maintenance stops more failures than reactive maintenance, it too has a downside. Unless preventive maintenance is properly optimized, an asset can easily be over-maintained (leading to overruns on maintenance and excessive downtime) or under-maintained (leading to costly firefights when an asset fails).

Condition-based maintenance is ideal when the conditions leading to asset failure are easily understood and measured. These solutions use rule-based logic to trigger alerts whenever an asset operates outside of defined parameters.

For those assets where failure conditions are harder to correlate and where asset downtime is costly and best avoided, predictive maintenance is ideal. This maintenance strategy uses advanced pattern recognition and machine learning to capture abnormal behavior that is often hidden from a cursory review of operations data. These revelations typically occur long before a normal alarm would provide notification, so predictive maintenance is best for high-priority, mission-critical assets where failure and downtime are costly.

How predictive maintenance works. Reactive and preventive maintenance alone are not enough to ensure optimal asset performance and reliability. Refineries are extremely complex, and equipment malfunctions can cause devastating failures. The downstream effects of refinery failure could have a negative impact on a company’s bottom line and reputation, as well as on the environment. With such high stakes, especially for mission-critical assets, oil and gas companies are increasingly taking a more proactive approach.

Predictive maintenance solutions leverage big data already being collected by oil and gas companies from historians, sensors and smart/IIoT devices. These solutions monitor the asset in real time to determine an asset’s base operational state. Advanced pattern recognition and machine learning are used to analyze each asset’s data to detect any abnormality or anomaly that might be a precurser to future asset failure.

Predictive maintenance and digital transformation. Predictive maintenance achieves impressive benefits, even when deployed in isolation. However, even greater value can be extracted by taking a more holistic perspective across the enterprise. In the best-case scenario, companies can install an industrial software platform, applying components of that platform as a systematic approach, where and when they add new value.

For example, once a predictive maintenance solution has flagged an issue, it is then critical that the issue is resolved quickly and efficiently. Integration with an advanced enterprise asset management (EAM) solution ensures that problems are actually resolved in an effective manner and are catalogued appropriately.

The produced insights and actions can be communicated to workers through mobile operator rounds and enforced as best practices in the field. Integration with augmented reality (AR) and mixed reality technology empowers workers with in-depth knowledge of the asset directly at their fingertips, facilitating operations and maintenance.

The benefits of digital transformation extend beyond asset management. By integrating asset management systems with advanced engineering and simulation capabilities, operators can create and maintain a complete asset history, from design through construction, operations and maintenance, adding another layer of information to users during asset operations and maintenance.

Improving asset reliability has valuable implications on the operations side, as well. By improving asset reliability, companies can improve the reliability of every connected process. Besides the financial benefits of preventing asset failure, improving asset reliability can also provide a boost to the environmental, regulatory and sustainability processes that come into contact with the asset, helping boost profitability and safety.

The industrial software platform in action. With the industrial software platform, companies can take a longer-term approach to the optimization of asset performance and operations. Asset lifecycles can be extended, helping drive the most from these investments to maximize return on capital.

For example, Hydrocarbon Processing covered BASF’s digital transformation.1 The company installed predictive maintenance and implemented AR technology to transform its work process. The new mobile solution is replacing paper checklists that operators formally used to conduct rounds, capturing data directly from inspection for real-time visibility. 

Overall adoption rates of predictive maintenance and APM technologies in the oil and gas industry are starting to increase. Early adopters are experiencing success, and with an ROI measured in months, not years, the only question is: When will you begin your digital transformation? HP


  1. Hydrocarbon Processing, “Schneider Electric Innovation Summit 2017: The transformative power of digitalization,

The Author

Related Articles

From the Archive



{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}