Andreas Gerndt

During the design of systems, engineers collaborate across disciplines with the aim of identifying a suitable design that matches their expectations. Such design is composed of a group of components, called a configuration. A configuration is Pareto-optimal if there is no other configuration that improves at least one objective without worsening another. Identifying such configurations is often performed using processes that are separated from one another. This costs time and is error prone during their integration. Despite contributions in this field, engineers are still provided with Pareto-optimal configurations but no means of analyzing their impact across coupled subsystems. It remains difficult for engineers to decide which configuration to select for their design. Moreover, tradeoff analysis that enables the evaluation of alternative configurations is usually performed using separate tools. This increases the burdens for engineers who understand the changes across subsystems, but have to evaluate these changes to find an optimal configuration. To address these challenges, this paper presents Change Impact Analysis Tradeoff (CIAT), an approach that identifies and analyzes configurations dynamically. To achieve this, it integrates knowledge graphs with change propagation, impact analysis, and tradeoff analysis in a single process. These capabilities favor dynamic analysis of configurations that can later be used by a digital twin to support design decisions. The proposed approach is currently illustrated in the context of space system design.

The increasing complexity and scale of scientific datasets demand advanced tools for efficient discovery and exploration. Traditional search systems often fall short in addressing the multidimensional nature of data and their inherent relationships, limiting their effectiveness for exploratory search. This paper presents the visualization-enabled search application (VESA), a backend-agnostic visual analytics framework that supports interactive and multidimensional exploration of heterogeneous scientific metadata. VESA enables cross-repository data discovery through configurable adapter interfaces that map repository-specific metadata onto a minimal set of common discovery dimensions, enabling seamless ingestion and consistent processing. At the frontend, coordinated visualizations, including spatial, temporal, and relational views, support exploratory search and sensemaking. To demonstrate the functionality and usefulness of the system, a software prototype is developed and applied to Earth System Science repositories, showing how heterogeneous sources can be integrated and explored through a unified interface. The framework is evaluated against established design guidelines and further validated through an online user study. In addition, adapters for two data repositories are implemented, illustrating how different backends can be connected to the system. Results indicate positive user reception, highlighting VESA’s usability, low learning curve, and its potential to enhance data discovery workflows through interactive visual exploration.

Satellite radar altimeters provide crucial insights into polar oceans and their sea ice cover, enabling the estimation of sea level, sea ice freeboard, and thickness. These retrieval algorithms depend on accurate discrimination between radar altimeter waveforms from sea ice and ocean surfaces in heterogeneous and dynamic surface conditions. A further and less mature step is classifying different sea ice types in addition to the ice/ocean discrimination. We aim to develop new methods for a novel multi-category sea ice and ocean surface classification directly from satellite radar altimeter data to improve sea ice climate data records. Traditional waveform representations are limited to a small set of parameters, leading to information loss. Moreover, machine learning models for sea ice classification often depend on supervised training, which is vulnerable to uncertainties in labeled data, especially in polar regions. To address these limitations, we explore self-supervised learning methods to optimize waveform representations, which can capture more detailed information for a classification with finer granularity. Furthermore, they do not require labeled data, which is not available at the spatial coverage and resolution of radar altimeter waveforms. We apply these techniques to SRAL data from the Sentinel-3 mission. We show that the information preserved in the latent space of an auto-encoder enhances the feature space of traditional waveform parameters, improving the subsequent classification process, when comparing our results to available sea ice charts and other remote sensing products. Our results demonstrate better generalization compared to supervised approaches.