Wintershall Dea always has an eye to the future, as is evidenced by its commitment to technological innovation, employee empowerment and environmentally responsible energy production. Recent events have made that foresight more important than ever.

The current iteration of the corporation formed in 2019 as the result of a merger between two legacy companies, Wintershall and DEA Deutsche Erdoel AG, each of which had been in business for more than 120 years.

As the two companies joined forces, the need for AI to connect and capitalize on data from across the larger organization became clear. Not only could AI provide greater efficiency and cost savings, but it also could bolster Wintershall Dea’s position as an industry leader in technological innovation—an important factor in attracting collaborators inside and outside of the organization. 

By 2020, the newly formed company was ready to accelerate an AI@Scale initiative. Several internal business and corporate units were already working with AI at that time, but the projects were developed for single, isolated purposes. With an AI@Scale approach, AI projects are built for scalability from the beginning. If successful, they can be expanded upon and extended to other groups rapidly and easily.

Developing AI@Scale solutions requires a centralized platform and methodology. “We want to be seen as a partner of choice,” says Kathrin Dufour, Senior Vice President of Digitization and Technology at Wintershall Dea. “If you have a standardized environment through which you can access your data and grant access to others, it becomes much easier to collaborate within your partner ecosystem. This is increasingly important today, because we exchange data both internally and externally much more than in the past.”

With that objective in mind, the company has established a center of competence (CoC) for AI and data science under the leadership of Ulrich Lorang, who serves as the Vice President of Data Science, Data Governance, and Data Hub. For planning, developing and delivering its AI CoC and platform, Wintershall Dea needed to team with a partner that could provide access to a larger and broader reservoir of AI expertise and experience.

In the search for that partner, IBM Consulting® stood out from competitors: It had a proven track record from working with Wintershall Dea on previous projects, as well as extensive experience helping other clients develop their AI capabilities at scale.

Additionally, IBM had formed a strategic partnership with Microsoft, and Wintershall Dea was already using Microsoft Azure for its data platform. IBM was able to adapt its IBM® AI@Scale methodology to accommodate the existing platform and to bring in Microsoft expertise as needed.

From the beginning, collaboration between IBM and Wintershall Dea was seamless. “The process was actually very simple,” says Lorang. “We built one team. There was never a big differentiation between the two companies. We had a common goal and worked together to fulfill it.”

In implementing IBM AI@Scale, the team focused on three strategic areas: the technical architecture of the platform, the operating model of the CoC and the corporate culture.

The IBM AI@Scale offering includes standardized assessments covering such areas as the current state of AI within a company, the future AI vision, key stakeholders and necessary resources. IBM customized the assessments for Wintershall Dea to cover each of the three strategic areas. IBM and Wintershall Dea then worked together to use the assessment results to develop a technological and organizational roadmap for data science within the company.

For the technical foundation, the team developed a component-based architecture using the Microsoft Azure platform and services. In designing this foundation, the team adopted a machine learning operations (MLOps) methodology—an end-to-end approach that taps into data scientists and engineers to plan, develop, build, test and maintain AI systems.

Operationally, the team laid out how the CoC should function, as well as the types of roles and skills that would extend data science capabilities across the organization. In addition to the data scientists in the CoC, that community included citizen data scientists from the business and corporate units—geoscientists, engineers, economists and others with strong mathematical programming backgrounds—who could help drive data science projects within their respective teams.

Ultimately, Wintershall Dea wanted to grow this community, upskilling employees throughout the company so they could develop their own AI projects. Interest in data science was strong, and management believed most of the value for the company started at the business and corporate unit level. To that end, the roadmap included technical enablement sessions for the CoC and the citizen data scientists on how to use the new platform and templates.

From a company culture and communication perspective, the team planned a variety of educational sessions and workshops for business and corporate units throughout the company. These activities focused on the business value AI could provide employees in their daily jobs and ways they could work with the CoC to capitalize on that value.

In 2021, IBM and Wintershall Dea were ready to begin laying the groundwork for the AI@Scale implementation based on the roadmap they had co-created. Setting up the technical environment, talking with the business units, identifying possible use cases, promoting the concept of the CoC, and engaging and enabling employees were all part of that process.

On the technology front, the team proceeded to provision the necessary services out of the Azure platform and to customize IBM AI@Scale templates to Wintershall Dea’s needs and environment. From an operational perspective, the team conducted technical enablement sessions to empower the CoC and citizen data scientists in the business units to create their own AI projects in the future.

To build employee awareness of the value of data science and the CoC—from both technological and business perspectives—the team met with the business units individually, held educational sessions and developed an internal campaign promoting the potential benefits of AI and explaining how the CoC could help.

In 2022, the team began the next phase of its AI journey: developing use cases into full-fledged solutions. The process for selecting and developing use cases followed along the lines of the IBM Garage™ methodology. IBM and Wintershall Dea worked side by side, with IBM providing guidance throughout the process, educating and enabling Wintershall Dea employees so they could duplicate the methodology in future projects.

The qualification process involved tight collaboration with the business units to understand their issues. “We worked closely with the domain experts to make sure we were not automating something just because we could, but we were really keeping the business problem in focus,” says Max Schemmer, Research-Oriented Artificial Intelligence Consultant of IBM Consulting.

Lorang concurs: “You must have a business problem. And you need to understand the challenges in your area and make sure you have access to high-quality, relevant data and then prepare the data so you can actually do something with it.”

Wintershall Dea primarily conducts two types of AI projects: traditional, large-scale projects and small, easy-to-implement “fireflies.” A firefly is a Wintershall Dea concept for conducting a quick, scalable AI project to solve a simple problem. Since there are employees trained in data science throughout the company, business units can develop and code fireflies independently and call upon the CoC for support as needed.

Fireflies start small, then sometimes catch fire. When they do, they are built to scale quickly. For example, an employee in an engineering department was tasked with manually extracting key values from more than 2,000 PDF documents and feeding that data into a spreadsheet. The process was tedious and took time away from the employee for more creative, meaningful work.

Applying AI, the engineering team was able to automate the process, enabling the employee to work on more challenging projects and providing greater overall value to the company. It soon became apparent that the same model for extracting real-time data from internal and external sources could be valuable to other parts of the business and beyond. Today, the scalable solution is applied in several business and corporate units for a variety of purposes.

Large-scale projects aim high from the beginning. In 2021, Wintershall Dea investigated the application of AI to maintaining the integrity of its gas and oil wells in Norway. That maintenance is particularly important for wells in operation, especially subsea wells. With miles of massive pipes encased in multiple layers of steel and concrete burrowing deep into the seabed, small leaks could be imperceptible for long periods of time even in the presence of state-of-the-art well monitoring system—until they become large enough to cause major problems in the worst cases. Thus, early detection is essential.

Previously, Wintershall Dea engineers had been monitoring data from well sensors on an ongoing basis. But even with day-to-day analysis, some issues were simply indetectable to human beings.

Using AI, the team developed a use case for analyzing data from existing sensors much more intensively and accurately than was previously possible. “We first sought to validate the hypothesis that we could use AI to detect a historical leakage incident,” says Prihandono Aditama, Product Manager at Wintershall Dea. “Once we were able to confirm we had the right model, we connected it with live data from the well sensors.

“Currently, if the AI detects an anomaly, it sends an email to our engineers,” he continues. “We’re in the process of building a user interface for the engineers, which will be available in the first release of the product.”

The IBM AI@Scale tools and methodology have been instrumental throughout the process. “One of the major contributions IBM brings to us is how to take the proof of concept live into production,” says Aditama. “The templates IBM provides enable us to do quick scaling, and to do testing, proofs of concept and development in parallel.”

Today, Wintershall Dea, together with IBM, has identified more than 80 possible AI and data science use cases, 20 of which it is actively pursuing. The use cases reach from technical areas, such as operations, engineering and geoscience to non-technical, such as commercial and sales. IBM is heavily involved in several of those, but the others are independently run within the company business and corporate units and the CoC.

Existing projects are progressing. The Wintershall Dea team is working on scaling the PDF extractor model to other parts of the company, pulling and applying data from internal and external databases. The well integrity project has launched into production status in late 2022. Post-launch, the team has preliminary plans to scale it both vertically—adding new features and capabilities—and horizontally—applying the model to additional wells in Norway and other countries.

Enthusiasm throughout the company for the data science initiative is high, both in terms of its potential to solve business problems and opportunities for innovation and skills growth. Over 100 Wintershall Dea employees have undergone AI and data science training, including 60 employees who attended a recent six-day data science workshop.

“We definitely have inspired the organization,” says Lorang. “We have built on a citizen data science community that is engaged and working toward using AI to solve problems with our help.”

The relationship with IBM continues to be strong. “We have a productive collaboration with IBM Consulting. Our joint efforts have generated significant momentum, enabling us to reach crucial milestones and deliver value in a relatively condensed timeframe,” says Hugo Dijkgraaf, CTO of Wintershall Dea. “Not only have they brought in AI skills and experience, but also the team has personalities well matched to ours.”

Wintershall Dea (link resides outside of ibm.com) is one of Europe's leading independent gas and oil companies headquartered in Germany. Formed in 2019 from a merger of Wintershall Holding GmbH and DEA Deutsche Erdoel AG, the company operates in 11 countries and has approximately 2,000 employees.