The state of the oil & gas (O&G) industry is stabilizing, as the US rig count is steadily increasing from its lows and the West Texas Intermediate (WTI) crude oil price has hovered around $50/bbl since its bottom in early 2016. Still, the industry has not rebounded to the peaks it saw in 2014. While many from the consumer side are happier to see the price of oil staying in the $50/bbl range, the upstream exploration and production (E&P) sector will need to adjust to these consistently lower prices.
(See Figure 1)
The new normal of prices staying lower for longer is having a ripple effect throughout the industry. The difficult pricing conditions for operators will put downward pricing pressure on service providers who are often at the end of the pricing bullwhip effect. Offshore operations will continue to operate on the margin of viability and will have a much slower recovery. Unconventionals will continue to play a pivotal role as the swing producer (a producer that can increase or decrease commodity supply at minimal additional internal cost, thus influencing prices and balancing the markets). In general, now that the pricing shock from 2014 to 2016—where the price of WTI declined from $100/bbl to $25/bbl—is over and the industry can see some stability, oil and gas providers can now focus on strategies around the long-term macro challenges they face.
(See Figure 2)
The companies that have made it through this most recent downturn and slow recovery still face many issues.
- A Changing Workforce
Many experienced workers have already left the industry, and those who remain are close to retirement. Although younger workers bring new skills and energy, they lack the experience and knowledge of those they are replacing.
- Lower Oil Prices
Unconventionals have become the marginal producer. Barring geopolitical issues, the rise of US shale producers has increased supply and thus created a price ceiling.
- The continued emphasis on health, safety, and environment (HSE)
These challenges compel O&G companies to find new ways to enhance the capabilities of their changing workforce, build more efficient operations, and minimize their impact on the environment, all while continually improving health and safety practices.
While there are disparate technologies that address each individual challenge, is there a technological solution that can handle them all?
Artificial intelligence (AI) and its most promising subfield, machine learning, have been around for decades, making promises that, quite frankly, have never been delivered. So, what’s different now, and why should O&G take notice of what AI claims to offer?
A SEA OF DATA
Rapid drops in sensor and data storage costs over the past decade have resulted in a dramatic increase in sensor data measured and stored for industrial equipment. This vast data can provide tremendous insight—if you can make sense of it. In addition, affordable, high-speed connectivity to remote locations where drilling activities typically take place is becoming more commonplace.
Significant increases in computer processing speeds, coupled with new platforms for handling large amounts of data and improvements in algorithmic techniques, allow AI to start solving real world problems (See Figure 3). Evidence of this is all around us in the consumer markets: Google’s autonomous Waymo vehicle, Netflix’s recommendation engine, Facebook’s facial recognition, and many more.
While the younger generation coming into O&G doesn’t bring much industry experience, they do bring a comfortability with data and connected devices, as well as an expectation of their presence. Rather than feeling threatened by a computer or software program which could potentially take their job, they have no qualms working side by side with hardware and software that will help them do their job better.
These three conditions combine to create an ideal environment for AI. The sea of data has grown beyond human capabilities to process. Rather than using pre-written rules or physics-based programs to tell the computer what to do, machine learning thrives on large data sets by building models based on observed data.
With the advancements in AI and the macro trends leading toward the need for machine learning, the O&G industry is poised to use this technology to solve its most pressing problems.
A CHANGING WORKFORCE
AI systems can capture the knowledge and decisions of experienced workers and institutionalize it for less experienced and future workers. For example, if a production engineer is monitoring a well, the previous approach would have been to spend hours looking at each well’s characteristics to determine if it requires maintenance. A machine learning model can look at all the wells that previously needed maintenance and identify a more advanced pattern of data that led a production engineer to make a certain maintenance decision. Then that model can be applied across thousands of wells in a field, flagging the wells that are at higher risk of needing maintenance. Here, the AI model has learned from the experienced production engineer and extended that learning to less experienced employees. Additionally, it allows the less experienced engineers to quickly evaluate a larger number of wells and focus only on the wells that need the most attention.
The need to eliminate non-productive time (NPT) is an important objective of many operators. The main cause of NPT is unplanned equipment downtime, which can be addressed by the advanced predictive analytics capabilities of AI. These capabilities can provide days to weeks of failure forewarning for critical assets such as artificial lift systems, top drives, draw works, frac pumps, compressors, and more. Having this amount of advanced warning allows operators and service companies to get the right personnel and equipment on site to fix a problem before it occurs, or to plan maintenance ahead of time and amortize the cost of that maintenance over many assets.
HEALTH, SAFETY, AND ENVIRONMENT
The desire to protect people and the environment is a top priority for all E&P operators. AI can monitor assets to ensure safe operations. Critical equipment (such as blowout preventers and pipelines) and the structural integrity of offshore platforms can all be monitored to identify data patterns that alert of impending safety issues. Furthermore, AI can monitor drilling conditions to detect issues such as well kicks.
The oil industry is now stabilizing from one of the most severe downturns it has ever seen. Although focused on long-term sustainable growth, the industry faces major challenges—a changing workforce, lower margins, and HSE concerns—all of which must be addressed with technology. The simultaneous rise in big data, computing power, and a data-friendly workforce has enabled AI to take the role of the all-encompassing technological solution the industry needs.
Last modified: November 22, 2017