A Capability-Based Framework for Classifying and Evaluating Digital Twins in the Oil and Gas Industry

Digital Twin technologies have seen growing adoption in the Oil and Gas industry, yet the term is applied to systems with markedly different levels of data integration, analytical capability, physical–digital interaction, and operational autonomy. Existing classifications distinguish broad categories such as Digital Models, Digital Shadows, and Digital Twins but provide limited guidance for the Oil and Gas context, as they do not explicitly differentiate behavioral representation, analytical sophistication, asset-to-model synchronization, digital-to-physical interaction, and decision-making autonomy. This limitation makes it difficult to compare implementations that share the same Digital Twin label but differ substantially in functional configuration and sophistication. To address this gap, this paper proposes a framework comprising seven capability-based categories: Static Digital Model, Dynamic Digital Model, Analytic Digital Model, Digital Shadow, Analytic Digital Shadow, Digital Twin, and Intelligent Digital Twin. The framework was developed through research synthesis and descriptive analysis and empirically assessed by classifying 92 industry-specific studies using a hybrid procedure that combined Generative AI (G-AI) assessments with researcher review. The results show that many systems labeled as Digital Twins in the reviewed literature are reported at intermediate levels, while incomplete reporting of synchronization, analytics, and decision autonomy contributes to classification ambiguity. The proposed framework clarifies the boundaries among Digital Twin configurations and supports the evaluation of existing applications, the identification of capability gaps, and the development of more integrated and autonomous solutions.