Congratulations! Lars Nonnemann, an employee in the "Visual Computing" department in Rostock, successfully defended his dissertation titled "A Multifaceted Model for the Layered Orchestration of Independent Visual Analysis Tools" on June 30, 2025.

The growing amount of information we generate and collect daily requires new, sophisticated solutions to facilitate analysis and gain insights. For this reason, new visual analytics tools are constantly being developed, tailored to specific application domains. These implementations range from descriptive libraries to fully-fledged frameworks designed to support the workflows of domain experts.

However, the heterogeneous and isolated nature of these tools poses a challenge when different data sources and functionalities need to be combined. Current methods for coordinating tools typically involve either developing a new system that integrates all functions or requiring users to manually apply different tools at various stages of analysis. Both approaches are resource-intensive and cumbersome, limiting usability and efficiency for domain experts.

To provide a more flexible solution, this work presents a novel approach to orchestrating visual analytics tools through a layered, workflow-oriented coordination model. This model aims to overcome the conceptual, spatial, and temporal separations of interactive visual analysis that currently hinder the effective use of multi-tool workflows.

Through a lightweight coordination model, this approach enables flexible, minimally invasive coupling at the data level between otherwise independent tools. Complemented by a well-developed approach for view-level visual integration (VI) ensembles, this work thus offers a practical solution to the mantra "Integrate what is necessary and couple what is possible," in order to meet the domain-specific needs of users.

The multilayered model was operationalized through both a configuration and a control interface and implemented as an open-source application. This application was developed and optimized in a three-stage evaluation process together with various user groups, to support user-oriented creation and execution of workflows in the form of analytical toolchains.

Step-by-step evaluation confirmed the feasibility of the approach in real-world scenarios—particularly in medical data analysis, where the efficiency and usability of workflows were improved by aligning them with users' actual analysis processes.

By linking independent visual analyses through a unified interface, this work lays the foundation for sustainable, adaptable processes involving multiple visual analytics tools and promotes the transition to more accessible, user-centered, and integrated solutions in the field of visual analysis.