This thesis addresses these challenges by proposing an architectural framework for PDTs based on the Model–View–Controller (MVC) design pattern. The framework enables a continuously evolving and adaptable digital representation of operational business processes that integrates process data, performance analysis, cross-system context, and intervention capabilities.

Following a Design Science Research methodology, the proposed framework is implemented as MVC-PDT, a working PDT prototype that supports process owners in integrating multiple process views from business systems such as enterprise resource planning (ERP), customer relationship management (CRM), and IT service management (ITSM) systems. MVC-PDT further leverages large language models (LLMs) to contextualize and explain process insights across these views. The collected process data and LLM-enabled contextualization are synchronized within a unified environment, allowing process analysts to identify optimization opportunities and initiate interventions directly in the underlying business systems.

The quantitative evaluation demonstrates that MVC-PDT supports real-time process data ingestion, integrates process insights from across heterogenous and enables immediate process interventions. Complementing these findings, a qualitative study indicates that users find MVC-PDT useful, intuitive and helpful while identifying real-time process monitoring and cross-system alignment as important enablers of process governance and optimisation.