As oil prices have slumped, North Sea oil production costs have risen and production efficiency has continued a steady decline -to levels where projects have had to be shelved and activity postponed, challenging the viability of existing fields and infrastructure and putting in doubt potential for resource recovery in the region. A study by McKinsey, a management consultancy, has identified an “asset production efficiency crisis” (together with other factors such as production decline rates as fields get older) as one of the main culprits.
While the North Sea has historically been one of the most cost-efficient areas in global oil and gas, lifting costs have tripled over a 15-year period. Given global price conditions, McKinsey estimates that a 75% reduction in cost inflation and production decline rates is required for unit costs to look sustainable in the longer term.
Oil production in the North Sea has matured and it now is a highly complex ecosystem with more operators, active platforms and producing fields than ever before. Safety and environmental regulation are more extensive and tougher than a decade ago. Higher activity levels have put pressure on the supply chain and led to higher costs, including higher factor prices –though suppliers don’t seem to benefit from the price increases as their margins are either constant or, in fact, reducing.
A deeper analysis by McKinsey of cost and activity data reveals that the increase in total expenditure often conceals cost inefficiencies in both OPEX and CAPEX and suggests that many operators have lacked the capabilities to combat cost escalation. Part of the increase is due to higher than expected maintenance (OPEX) and modification (CAPEX) requirements as installations have increased in number and aged to well beyond original design life. This in turn drives additional logistical activity (transport infrastructure, materials tonnage). It also drives increases in prices (due to supply constraints) for supporting services, equipment, consumables and skilled personnel. Another part of the increase is due to unexpected lower efficiencies and reducing productivity, measured across a broad base of indicators, chiefly the result of operator practices and approaches. For example in 2012, 179 core personnel travelled offshore per manned installation –a 26% increase from 2006. Work productivity (hours per activity – chiefly O&M activities) declined 4% p.a. between 2001 and 2009. At one installation drilling speeds has dropped from 175m per day in 2004 to less than 100m per day in 2014.
Data from the UK Department of Energy and Climate Change confirm that production efficiency (defined as percentage of estimated production potential) in the UK-side of the North Sea declined from an average of 81% in 2004 to 60% in 2012. The losses in potential production led to government involvement and the creation of joint government-industry bodies like PILOT and its Production Efficiency Task Force to analyze causes and find solutions. While the causes are complex (including age, installation interdependencies and bottlenecks) McKinsey nevertheless estimates that a significant reason lies in asset management practices –specifically operator maintenance and reliability practices and approaches, the quality of which correlate highly with production efficiency on both individual field and operator level.
Figure 1 (Source: McKinsey)
McKinsey analyzed the differences between operators with high quality and those with poor asset management practices and found three critical differences, namely that good operators:
- Challenge and minimize planned downtime
- Continually improve reliability by learning from failures
- Create a culture of responsibility in operations
Challenge and minimize planned downtime
This is an interesting finding, as the default position usually is to minimize unplanned downtime while treating planned downtime as a good time to complete many maintenance tasks, including those that have been building up since the last intervention. McKinsey found however, that operators performed better when only the most essential activities were carried out and everything else possible performed during normal operations. Of course this speaks for moving from preventive to predictive types of maintenance and tools within broader RCM programs to reduce perceived required maintenance amounts and activities, not least by avoiding new requirements caused by the interventions themselves.
Continuously improve reliability by learning from failures
The study found that high performers were very thorough in conducting in-depth investigations into root causes of failure that resulted in lost production often making use of advanced analytics techniques. Once causes were identified these companies followed through by making appropriate changes to equipment, operating protocols or maintenance strategy to minimize probability of recurrence. The investigations root out causes of losses both from larger events and frequent smaller failures.
Create a culture of responsibility in operations
This is perhaps the most important finding, also because it is intricately linked with the previous ones, in the sense that it enables or reinforces the principles: The best performing operators have been very successful in instilling a real sense of responsibility for the care of equipment by those who use it every day. They use the term “ownership” to describe attitude and behavior of employees. This culture of responsibility and “ownership” is supported by widespread adoption of operating standards. They not only maintain written standards that minimize the chance of equipment damage during production, they also make sure that standards are visual, accessible and written in plain language and that management processes are in place to identify and manage deviations from those standards –as part of normal every-day work for supervisors and managers.
And in an interesting take on maintenance outsourcing McKinsey notes: “It is this final point on the importance of creating a sense of responsibility that has been frequently cited as the reason many operators have struggled to make the outsourced operations and maintenance model work well. While good performance can be achieved with outsourced offshore activities, doing so requires a clear delineation of responsibilities and a careful alignment of incentives with the service provider. This has proven difficult, and our data suggest that when this is not done well, production efficiency performance can suffer”.
We have dealt with the issues of delineation of responsibilities and alignment of incentives in our Brief Outcome Based Services and Performance Based Contracting. And we will come back to this important topic in the future, as it is indeed the case that it is frequently one of the main reasons of why outsourcing and outcome / performance based contract don’t live up to expectations.
In any case, while the McKinsey analysis was undoubtedly thorough and the findings and conclusions drawn were correct –though hardly surprising – as far as asset management practices go, it is important to note that the deeper question was actually not asked, namely why, given the high correlation between asset management practices and production efficiency and the fact that production efficiency was originally high, did asset maintenance practices deteriorate in the first place? Because this, though not directly evident from Figure 1, is what must have happened – in spite of improvements in both asset management methodologies and technologies that have occurred over the past 10 years or so.
One explanation may lie in the high oil price that prevailed over most of the past decade – the high margins may have caused operators to slacken their focus on asset productivity and cost containment. This would have meant however that production efficiency should have increased by now, given sustained current low prices and margin pressures.
Another explanation however, which appears more intriguing, is that complexity in North Sea oil production has increased so much that management, including operations, maintenance, asset management or supply management have become really difficult – in the sense that a marginal increase in complexity results in a proportionately greater requirement in management effort –other things being equal. In other words it is necessary to run in order to stand still.
The growth in complexity can be proxied by various measures, but in essence is determined by the amount of activity, information flows and decision making taking place concurrently in a limited geographic space, while following an increasingly rigid and voluminous regulations regime. Interdependencies and negative feedback loops increase exponentially to the point where diminishing returns become hugely significant. If this assumption is correct, then the clear focus has to be on –apart of course from continuous improvement in practices- :
- Reduction in complexity by ruthless elimination of all kinds of unnecessary activities –not only in maintenance
- Investment in technology and systems that are better able to manage complexity, which brings us of course to the Internet of Things and data analytics. It appears that difficult to manage highly complex environments should be a high priority application for this technology which provides potential for real time understanding of interdependencies and process self-regulation
The North Sea region has always been a frontrunner in service thinking, asset management methods and performance based business models. It is now entering a new maturity phase and developments there will both be interesting to watch and will influence what happens elsewhere.
Join the Community
We are building a community of service in industry professionals -business leaders, management practitioners, digitization experts, technical experts, innovators, technologists, consultants, academics, and investors.
Join our community to receive articles, briefings, guides, news analysis and more.
Deep dive into the industrial service business.
Join our community to receive analysis, insight, news and more.
We will never share your data