Impact case study database

Context

Impact has been created through the commercial exploitation by energy industry multinational Schlumberger of the AI planning tools developed by Fox and Long. This has resulted in a step change in the technology applied to automatic drilling operations, with economic, efficiency, safety and environmental benefits.

Schlumberger is the world's leading provider of technology and digital solutions for reservoir characterisation, drilling, production, and processing to the energy industry. The company supplies an extensive range of products and services, from exploration through production, with a major focus on delivering hydrocarbon reservoir performance sustainably. In terms of scale, the company has product sales and services in more than 120 countries, employed approximately 82,000 people (representing over 170 nationalities) as of the end of third quarter of 2020, and reported revenues of USD32,920,000,000 in 2019.

From research to impact

Schlumberger became aware of Fox and Long’s work on planning with continuous processes through the company’s participation as an industry partner in the EPSRC Autonomous Intelligent Systems programme in 2011. Their research has subsequently been instrumental to the company’s introduction of autonomous directional drilling [ S1]. After initial evaluation, the company engaged Fox and Long as consultants from 2013, during which time they assisted with the development of an architecture for the later adoption of their methods in the application of automated drilling [ S1]. Both Fox and Long joined Schlumberger as employees in 2016.

The following impacts have been realised during the REF assessment period:

• Technology innovation, new product development, demonstration and validation;

• Improvements in well construction processes including time and cost savings, improved drilling consistency and reduced environmental footprint;

• Economic impacts including product sales and industry partnerships.

The development and implementation of AI planning supported Schlumberger’s strategic objective to move its services up through levels of automation, the scale of the reach and significance of the impact realised set out clearly by the company’s Vice President for Technology Development, Well Construction Division:

‘The real step change in our approach to automate highly dynamic and uncertain drilling process started when we engaged Derek and Maria first as consultants and eventually as full-time employees. Their work on the AI planner has become the core of our present and future automation products and services. We have established two large engineering centers, one in Houston, Texas and the other one in Beijing to engineer the research prototype that Derek and Maria developed into commercial products’ [ S2].

Impact 1: Technology innovation, new product development and demonstration

In the period 2013-2015, the POPF planner as developed by Fox and Long was trialled by Schlumberger and a demonstration system was built to explore the potential of using POPF to automate drilling in US land operations. In 2016-17, Schlumberger carried out field tests using POPF, with the intention of identifying the commercial potential of the plan-based automation of drilling.

Concurrently, Schlumberger began the construction of a commercial planning system under the direction of Fox and Long. This led to the development of a next generation planner using methods pioneered in POPF. In subsequent deployments, the new planner was used to generate the action sequences, timing, and parameters (collectively known as plans), needed to reach a given depth. The plans generated vary depending on the hole depth, trajectory, formations and other factors. Preliminary results from field deployments confirmed the potential of the plan-based automation method to speed up the drilling process and provide associated costs savings.

The underpinning technologies built on the POPF planning tool and VAL plan validation system are now embodied in the Schlumberger commercial DrillOps on-target well delivery solution [ S3] available within Schlumberger’s DELFI cognitive exploration and production (E&P) environment [ S4]. Also within the DELFI environment, the DrillPlan solution, which supports the initial well construction project planning phase, works in tandem with the subsequent automation of the drilling process via the DrillOps solution. The two solutions operate as a closely coupled pair in delivering increased automation in well construction.

Impact 2: Improvements to well construction processes: efficiency, consistency and reduced environmental risks

The implementation of AI based planning through the DrillOps and DrillPlan solutions represents a step-change in well exploration and construction. Previously, the industry standards were manual planning and a process known as scripting, in which a pre-written set of instructions, a script, inform the drilling process. Both processes come with considerable drawbacks: manual planning is time-consuming and vulnerable to human error, and scripts – although a significant improvement on purely manual planning – are usually written for a specific location, and thus lack the flexibility to be shared between drills at different sites. Both approaches can therefore lead to inconsistent drilling, delays, additional costs and errors, which may be dangerous or environmentally damaging.

Automation provides numerous benefits over both manual and scripting approaches. It results in greater efficiency and improved consistency of drilling operations. In contrast to scripting, which typically produces scripts specific to individual drill sites, automation is also a more flexible approach and thus applicable to a wider range of drill sites. This saves considerable time and allows engineers working in similar geological areas to share solutions to problems, resulting in significant efficiency improvements. In these ways the capability of automating plan generation has advanced the state of the art in automation techniques in well construction.

The transformative nature of the impact realised through the DrillOps solution is summed up by Schlumberger’s Research Programme Manager for Automation and Planning at the company’s Cambridge Research Centre:

‘The work has provided wide reaching business impact, such as through the DrillOps autonomous drilling solution, which is not incremental but marks a fundamental shift in the way we approach automation both within our own operations and in solutions we are developing with major oil and gas operators…Automation, through temporal planning enabled by the work of Fox and Long, creates consistency of operations, which drives safety, reduces time-to-target, and the minimisation of environmental footprint’ [ S1].

Specific operational performance improvements from proof-of-concept deployment in North America, the Middle East and Mexico include [ S5]:

• Improved drilling efficiency: a 20% increase in the drilling rate of penetration (ROP - the speed at which the drill bit can break the rock under it and thus deepen the wellbore);

• The ability to apply standardised procedures throughout a fleet of rigs was found to reduce non-productive time; tool failure non-productive time was reduced by 90% after the roll-out of the DrillOps solution;

• Analysis of drilling operations under manual and automatic control demonstrated that automatic control resulted in improved procedural adherence which has associated consistency and safety benefits, as well as reduced environmental footprint and risk.

Impact 3: Commercial implementation and associated economic impact

Since its commercial deployment started in 2019, the current generation of automated drilling products are used on 27 drilling land rigs [ S1], with each rig capable of drilling about 15 wells per year. The operational benefits outlined above are estimated to transfer into operating cost savings of around USD300,000 per well or USD4,500,000 per rig per annum [ S6].

In operational deployment, the step-change in operational consistency that the technology helps to deliver was reported in the Schlumberger financial results as a technical highlight of quarter 3 2020 [ S1]. This included reporting on deployment in Saudi Arabia where over 63,000ft had been drilled using the DrillOps solution, with a 17-30% improvement in rate of penetration [ S7].

In 2020, Schlumberger entered into commercial agreements with major global energy industry partners to integrate automation products including the DrillOps solution in Schlumberger systems developed for these partners. These include:

• Schlumberger and ExxonMobil, the world’s largest oil company, finalised an agreement allowing ExxonMobil to deploy the DrillOps solution as part of the deployment of digital drilling solutions around planning, execution, and continuous improvement through learning [ S8];

• Schlumberger and Chinese-based company Honghua Electric Co Ltd entered into a memorandum of understanding for the integration of the DrillOps solution into all new rigs manufactured by Honghua [ S8]. Schlumberger’s share value increased following this announcement.

Impact 4: Other applications of the planner within Schlumberger

The flexibility of the approach to AI planning, based in the Strathclyde underpinning research and embodied in the DrillOps solution, has allowed Schlumberger to implement it in other aspects of its services business [ S1].

Known as wireline operations, well measurement operations involve the deployment of sensors and associated transceivers via boreholes. These processes are typically operated manually, but this is accompanied by difficulties in controlling the winch, directing the cable and moving it a fixed speed. Schlumberger has successfully automated this process using the planner. As of 2020, Schlumberger’s system of automated wireline operations is currently under field test.

The logistics and product delivery requirements for well drilling are very large and complex, requiring a wide range of equipment and experienced staff, from a variety of locations, to all arrive at a drill site on a set day. Schlumberger has used its planner to successfully automate some aspects of this process in specific applications, with full deployment in 2020.

These new applications demonstrate the flexibility and broad applicability of the planner and its continued importance for Schlumberger’s operations.