After a merger, Wintershall Dea needed a scalable way to connect data, skills, and AI use cases across the larger organization because existing projects were isolated, difficult to scale, and not supported by a common delivery platform or operating model.
Reported impact
Wintershall Dea established a scalable AI operating model, identified dozens of viable use cases, trained more than 100 employees, automated document extraction work, and launched well-integrity AI into production with a stronger citizen-data-science community behind it.
Wintershall Dea partnered with IBM Consulting to build an AI-at-scale capability that combined platform architecture, operating-model design, and employee enablement. IBM adapted its AI@Scale methodology to the company’s existing Microsoft Azure data platform and used assessments to define a roadmap for a new center of competence, citizen-data-science community, and repeatable MLOps-based architecture. IBM then helped Wintershall Dea provision the technical environment, customize templates, run enablement sessions, and apply IBM Garage methods so business units could qualify and develop both small “firefly” projects and larger enterprise use cases. That framework supported solutions such as PDF extraction workflows that automate manual engineering tasks and well-integrity models that analyze live subsea sensor data for anomaly detection. Instead of treating AI as a sequence of pilots, Wintershall Dea created a structured platform for moving proofs of concept into production and for scaling successful ideas across operations, engineering, geoscience, and business functions.
Business and operational data from across Wintershall Dea feeds into the Azure-based AI platform, teams use AI@Scale templates and MLOps workflows to develop and test solutions, and resulting models are pushed into business-unit or production environments such as PDF extraction and live well-sensor monitoring.
“Applying AI, the engineering team was able to automate the process” of extracting key values from more than 2,000 PDF documents.
“IBM was able to adapt its IBM AI@Scale methodology to accommodate the existing platform”
“Over 100 Wintershall Dea employees have undergone AI and data science training”
“Today, Wintershall Dea, together with IBM, has identified more than 80 possible AI and data science use cases”
“the team developed a component-based architecture using the Microsoft Azure platform and services.”
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
IBM Sample Dec 3, 2025 CEMIG CEMIG needed to improve service quality and accessibility for millions of customers across digital and traditional channels, especially in rural and underserved areas where preferences, infrastructure, and digital… 1.1 million monthly WhatsApp messages Hugging Face · IBM Consulting · IBM watsonx Assistant IBM Sample Jul 29, 2025 Matrix Renewables Matrix Renewables needed to onboard and monitor a rapidly growing portfolio of solar assets without the delays, data gaps, and scaling limits caused by decentralized monitoring across multiple SCADA systems. 10% reduction in downtime IBM Maximo Renewables IBM Sample Jan 10, 2025 AGL Energy AGL needed to offer always-on digital support across web, account and mobile channels while reducing the load on frontline agents and enabling more transactions to be completed in self-service. 125,000 customers each month are starting a chat IBM watsonx Assistant 