Defining the digital revolution

Why the time is now for downstream operators. By Ron Beck

The downstream energy industry has reached a tipping point. Competition is fierce for a stagnant demand in Europe, prices volatile and margins thin as operators consider needed technology investments. All this is creating a logjam of capital investment, delayed by uncertainty of the marketplace future directions, and an urgent necessity for operators to deliver lean operations that maximise asset uptime and utilisation. Global oil and natural gas demand is shifting, as energy sources for transportation shift, leaving executives desperately searching for the key to applying digital technologies to react quickly and gain an edge on their rivals.

Digital transformation holds that key and operators need to grasp it if they are to survive and thrive in the future. There is a great opportunity opening up here that they can’t afford to ignore. Today, with the successes of AI, machine learning and multivariate analytics based energy solutions, early-mover companies have gained an advantage, addressing issues they’ve never previously been able to solve quickly enough, and at the same time attracting the best of the new generation of workers to the companies seen as exciting places to work and to take part in digitalisation.

Key areas where digitalisation is already helping refiners include extending asset life; maximising return on capital employed and driving additional profit. The ability to detect faults in advance of machinery and process unit breakdown can save vast sums. A compressor failure at an oil refinery, for example, can lead to downtime losses running into millions of pounds. Enough advance warning can yield operating changes to avoid the downtime; or planning and commercial changes to maximise results in face of unavoidable downtime. Given the challenges in the market today, operators simply can’t afford to be late movers in adopting these already valuable technologies.

Grasping the opportunity
The good news is that oil and gas companies are increasingly looking to embrace digitalisation. But in doing so they face a problem. How do they prioritise the multitude of digital initiatives they could get involved in without potentially reducing the chance of success?

Two significant areas present immediate opportunity. The first is autonomous dynamic optimisation. Following years of promise, tying together refinery and petrochemical economic planning with advanced process control in a closed loop process is possible. It is an approach that delivers both short-term economic benefits but also provides insights to push the limits of what’s possible.

The second area is prescriptive maintenance (based on machine learning), which gives operators the opportunity to reduce downtime across both their upstream and downstream operations. With this approach, embedded AI/machine learning provides real-time insight into plant assets, not only helping to determine weeks in advance if assets are likely to degrade or fail, but also, most importantly, providing information as to process conditions causing that degradation. That, in turn, helps make changes to improve uptime and operational efficiency for energy companies worldwide.

Operators must adopt a pragmatic approach to implementing these new digital technologies. That does not mean going after small problems or profit opportunities. Organisations can be pragmatic around a significant problem that represents major value.

Italian energy provider, Saras took this approach at its 300,000-barrel-per-day refinery in the Mediterranean, applying machine learning to four equipment areas: feed pumps, wash oil pumps, makeup H2 compressors and recycle compressors. It delivered its digital effort in a matter of weeks, able to accurately identify the specific failure mode for each component — without false positives in the same short timeframe.

These capabilities enabled Saras to predict failures with lead times of 24-45 days, and the refinery also expects to reduce unplanned shutdowns by up to ten days, increase revenue by one to three per cent and reduce refinery maintenance costs and operating expenses by one to five per cent1.

This project is one of a number of similarly successful implementations underway at refineries across the world. Now that companies are able to see and talk to implementors of successful applications of prescriptive machine learning, momentum is building in this area.

The numbers projected by several refiners range from two to seven additional days uptime per year. Depending on the size of the refinery, that incremental revenue is economically significant.

The time is now
These are times of opportunity for energy operators. The tools, services and solutions needed to overcome complexity and achieve new levels of reliability and profitability are now available, enabled by breakthroughs that make them accessible to any business.

By asking the right questions and targeting these digital technologies to their needs, oil and gas companies can apply them where they create most impact. This will help them achieve the highest possible return over the entire lifecycle.

The oil and gas industry has been talking about digitalisation for decades. But it is only now that enabling technologies and sophisticated machine learning and analytics algorithms have converged to tackle process degradation and equipment failure in real-time. This level of analysis opens a whole new area of value creation and reliability improvement for owner operators. The time is now to define the digital revolution.

1 “Prescriptive Maintenance Software Helps Saras Improve Business Performance and Drive Operational Excellence,” April 2018

Aspen Technology
Ron Beck is Strategy Director at Aspen Technology (AspenTech) a leading software supplier for optimising asset performance. AspenTech uniquely combines decades of process modelling expertise with machine learning. The company’s purpose-built software platform automates knowledge work and builds sustainable competitive advantage by delivering high returns over the entire asset lifecycle.

For further information please visit: