Energy companies are operating in uncertain times. They face societal pressure and increased regulation to significantly reduce fossil-fuel dependency, primarily characterized by reliance on transport fuels, plastics, and other refining and petrochemical products. Under these conditions, companies are looking to maximize the health and resilience of downstream operations—particularly in oil and gas and chemicals—by adopting new automation and digital technologies, enabling increased levels of data usage and performance transparency, as well as faster decision loops.
Many of these changes were already occurring and have only accelerated during the COVID-19 pandemic, resulting in a newfound sense of urgency. Yet digital strategies remain challenging for the energy sector for several reasons, namely keeping up with increased decarbonization efforts, workforce changes, and accelerated technological innovations. Many companies have responded with short-term solutions but remain indecisive about how to identify priorities in the years to come.
Autonomous plants are a promising solution. Such future plants link technology, data, and advanced visualizations with operations to ensure that assets learn from each action taken, as well as from historical data and derived insights. These plants react to asset health and economic conditions and progressively improve their operations over time to run with a lower carbon footprint as well as more safely and more profitably.
All plants, irrespective of their maturity levels, are primed to identify and adopt digital technologies to move toward autonomy.
Our research shows that all plants, irrespective of their maturity levels, are primed to identify and adopt digital technologies to move toward autonomy. The building blocks are now in place, the technologies are available, and the required skill sets are coming into focus. Actions taken today by executive leadership can maximize cash flow from downstream assets and move the plant toward renewable energy sources, thus helping meet the requirements and sustainability objectives.
The energy transition: Challenges and opportunities
Although the energy transition creates an imperative for companies to increase the resilience of their operations, it also presents strong headwinds for the industry. In fact, some companies have responded by naturally hedging against market risk through portfolio diversification, while others have redoubled their focus on achieving cost competitiveness through operational excellence and value-chain extensions (Exhibit 1). Either way, these companies seek to increase cash flow from operations to buffer against sustained market volatility, as well as to improve the attractiveness of assets that may be considered for divestiture.
As an example, one of the largest energy and chemical conglomerates in Asia recently addressed the uncertainty of the energy transition by embarking on a multiyear digitalization and automation initiative, with ambitious efficiency, sustainability, and competitive-advantage targets. Beginning with broad value-chain and digital-twin efforts that will immediately use digital technology to reduce production cost structure, reduce carbon intensity, and increase resilience, they now have a clear vision to deliver autonomous operations in the years to come, with incremental progress beginning in 2025. Such moves demonstrate how companies can take the initial steps to bolster autonomy and automation in the short term, with the aim of benefiting in the long term.1
The following trends highlight both the challenges and opportunities companies can expect to encounter as they embark upon their digital journeys:
Increased societal pressure and regulation continue to drive a shift in the environmental, social, and governance (ESG) imperative for downstream oil and gas and petrochemicals. Recent legislation, such as the European Green Deal, seeks to drastically reduce carbon emissions and increase economic sustainability. Such legislation drives increased investment in non-carbon energy sources and aims to reduce net demand for products derived from fossil fuels. It also incentivizes innovation and technology development. As a result, low-carbon and renewable energy sources are becoming economically feasible and thus more attractive. For the existing capital-intensive assets, the autonomous plant becomes an imperative to keep pace for decarbonization.
The oil price collapse in early 2020 disproportionately impacted highly experienced knowledge workers with crucial domain expertise. Now-familiar tropes include large-scale layoffs in both energy and chemicals; the early retirement of older, highly experienced workers; and a cyclical drop in the number of university candidates interested in joining asset-intensive organizations. The current market slowdown has accelerated the pace with which energy companies evaluate digital solutions to address the scarcity of technically skilled workers. And the previous year of remote work has demonstrated that achieving operational excellence is reliant on a combination of technology and technically skilled workers.
Accelerated technological innovation
As digital technologies continue to evolve and become concurrently more affordable and easier to deploy, the energy sector’s rate of adoption will continue to increase (see sidebar “A tipping point for digital technology”). Recent infrastructure upgrades, such as secure 5G site-level networks, have dramatically changed data-management capabilities and reduced cybersecurity risks, and they now allow for integrated, automated solutions to be deployed. In addition, these upgrades have significantly reduced the cost of connecting by network instead of hard wiring geographically disparate assets to control rooms, many of which are being moved physically farther away because of facility-siting considerations.
The autonomous plant: An overview
The autonomous plant incorporates recent technological advancements in connectivity and computing power—as well as access to Industrial Internet of Things (IIoT) data—in order to progressively improve performance based on data analytics, AI, and first-principle models without significant human intervention (Exhibit 2).
A carefully determined combination of conventional technologies, AI, ubiquitous data, connectivity, and collaboration coalesce to consider potential future states of refineries or petrochemical plants when making operational decisions. Prescriptive analytics, applied to the operating data and conventional models, offer technical personnel better information and can reveal alternative strategies, either advising operators or—where closed-loop systems are operational—taking autonomous control. This level of control is not only a key value driver but also enables knowledge workers to focus on the future state of the plant. By carefully combining technology and process, the increasingly autonomous can ensure that organizations possess a resilient and flexible set of assets that can react and be reconfigured to thrive in changing economic, operational, or demand-volatile markets (see sidebar “The full potential of autonomous technology”).
A fundamental element of the autonomous plant will be its ability to collapse and close the feedback loops between planning and scheduling and operational optimization technologies. This ensures that relevant insights are shared and that appropriate actions are identified and taken. As was made clear during the recent pandemic-induced slowdowns, many plants experienced challenges associated with running at minimum allowable rates during low-demand periods. In the absence of closed feedback loops, many plants were unable to orchestrate the multiple constraints and process units in a way that achieved production goals and maintained safety, as their planning and control systems were not designed to optimize at low throughputs. When running under so-called normal operating conditions, these closed feedback and feedforward loops also enable the plant to run closer to its limits in a safe manner, and, in some demonstrated cases, increase overall plant throughput limits by 5 to 10 percent.
The journey toward the autonomous plant can also elevate workers’ capabilities to make tactical decisions that are clearly aligned with strategic initiatives for improved operational integrity, sustainability, and production. Furthermore, the technology solutions can be designed to enable more effective collaboration between organizations and improved coordination during decision making.
A carefully determined combination of conventional technologies, AI, ubiquitous data, connectivity, and collaboration can work in concert to consider the future state of refineries or petrochemical plants.
Early adopters of key autonomous components have shone a light on the surprisingly high amount of value available through comprehensive implementations—or, by contrast, the amount of margin value currently lost by running ineffectively, from a systems point of view. For example, in refining, digital twins are often implemented to update planning models and monitor key equipment. These implementations are worth ten to 50 cents per barrel; dynamic multiunit optimization synchronized with planning, 15 to 30 cents per barrel; integrated planning and scheduling, 20 to 50 cents per barrel; and adaptive multivariate control, 12 to 20 cents per barrel. On top of that, the value of AI agents closely integrated with the digital systems can be worth more than another dollar per barrel. As the price of oil stays within the $60 to $80 window for some time, crack spreads would likely expand and observed benefits of digitalization would scale.
While all energy companies can benefit from increased autonomy, the level of benefit varies significantly depending on the underlying structural and macroeconomic conditions of the markets in which they operate. Individual organizations must use this information to temper both where they start their journey and how far they should travel. Depending on the strength of the headwinds the organization faces and the regional and local market dynamics, energy companies will need to determine their priorities based on their needs. With this in mind, four archetypes can help companies scale their investments accordingly (Exhibit 3).
These archetypes represent a spectrum: from rapid technology adopters, to companies evolving their business models, to those with a raw-materials advantage. Each archetype has a clear motivation for moving toward the future state of autonomous plants, with options that include simple control-systems upgrades to modular integration of bespoke digital solutions. Furthermore, as players focus on conserving cash flow based on market dynamics, it is important to highlight that many of these solutions can be deployed using “soft sensors,” or algorithms that optimize throughput and yield, rather than requiring significant capital investment.
As different plants can be at different levels of digital maturity, we mapped five distinct stages that apply to all archetypes, ranging from basic operations to the autonomous enterprise (Exhibit 4). This maturity model highlights the time needed to implement new technology and transform management systems, build workforce capabilities, and embed new behaviors. However, as levels of automation increase, there is a corresponding increase in value to the plant and enterprise ecosystem.
The spectrum of maturity ranges from basic operational-control systems at plant sites to fully integrated, autonomous companies. Each subsequent level represents thematic step changes in the levels of digitalization applied (see sidebar “Key questions for executives shifting to autonomous operations”). In our experience, the following examples help illustrate what companies at each stage may be doing:
- Basic digital adoption. With the simple application of control systems in operations, site-by-site basic refinery planning, spreadsheet-based scheduling, ad-hoc troubleshooting-induced process modeling, and maintenance work orders, this maturity level represents the widest adoption level today. The journey started here in the 1980s, but many organizations became mired at this phase. Companies taking this stage generally do not have any semblance of a digital organization and consider their next digital frontier to be improving their current systems to today’s best practices and some level of business-process collaboration.
- Selective advanced-analytics adoption. Five to ten years ago, a number of companies took the first step toward this next level, beginning with the limited implementation of APCs and the deployment of advanced linear programming–modeling techniques and enterprise resource planning (ERP) programs. The companies fully embracing this stage today are deploying advanced analytics for value-creating use cases, such as data analytics–based prescriptive maintenance for uptime, adaptive advanced process control, and online optimization of individual units to further orchestrate and improve value. While these are significant steps to take, individual leaders must still promote the piloting of digital solutions, as opposed to a comprehensive, cultural openness for disruptive technologies. Furthermore, companies have often neglected to develop a future-proof edge-to-enterprise sensor and data strategy. Finally, funding generally falls under the premise of “prove-it” trials.
- Cross-discipline optimization. Although much less common and typically exhibited only by today’s leaders, this stage sees companies exploring cross-discipline optimization for which multiple digital solutions are deployed and connected. At this point, these companies have generally solved the issue of scaling and therefore have a sense of where digital deployments can deliver value. Examples include companies with fully integrated planning and scheduling solutions, well-designed ERP systems that drive impact for the organization, multi-unit process optimizations, and asset-wide digital twins for emissions monitoring. They also have digital organizations, however nascent, as well as cultural excitement and momentum around what the technology solutions can do. A foundational step is attention to the edge-to-enterprise data strategy. If solutions are deployed correctly and the company experiences positive impact, digital programs often have become “self-funding,” and barriers to adoption lower significantly. This is an important stage to reach and—to prepare for the next maturity level—accelerated adoption should be a high priority today.
- Autonomous plant. Shifting to this stage requires significantly more trust in digital tools, organizational alignment around risk, and fully fledged digital organizations that drive continuous change. This means companies explore closed-loop process optimization, integrate workflows within sites, and likely experience overall cost reductions from reduced labor, increased process stability, and improved reliability. Front-line workers are firmly part of the culture of adoption and push technical support staff on how to continue the digital strategy. Although there are currently no autonomous plants, several sites are taking significant steps to deploy integrated solutions in that direction. Hybrid solutions that embrace AI are key at this step.
- Autonomous enterprise. An asset or plant only creates value in the context of the enterprise value chain. To unlock the full potential of the autonomous plant, the value chain must become intelligent. This means it can make decisions across the enterprise network prioritizing which assets produce which product and what objectives are being optimized for each product and production line at each site. As the name suggests, this step is the zenith of the journey toward autonomy. At this point, solutions are fully integrated, real-time data visualizations enable continuous cross-plant optimization, and site-level workflows are tied to equally advanced back-office workflows. There are essentially no companies that qualify as “autonomous enterprises,” although many are developing elements of an autonomous future. One example of this would be a company’s trading organization enabling real-time (or near-real-time) visibility into operations and logistics to continuously identify the latest relative value of crudes or the cost of producing various products to inform commercial decisions.
When thinking of the digital-maturity model, the key takeaway is that every organization has steps available to them—no matter their position or progress on the journey toward autonomy. The entire spectrum of plants and companies has opportunities to move forward to deploy digitalization and continue transforming their cost base. We recommend that companies first clarify their ambitions and then chart a path to whatever degree of autonomous operations is desired—but ultimately, the options are plentiful and often require limited or no capital investment. With continued adoption of digitalization across the plant and the enterprise, plants can become more autonomous and self-optimizing, laying the foundation for the autonomous enterprise.
Irrespective of digital-technology adoption, process-and-operations maturity, and geography, there are impactful steps that companies can and should take to improve the resilience of their plants and networks. Incorporating even the smallest advances can stimulate a shift in the ambition and energy of the enterprise. While often we hear the comment, “But we can’t afford to invest in technology in this environment,” we would assert that, in many cases, companies cannot afford not to.