ResourcifAI: AI-powered decision support for efficient environmental resource management
Project Description
ResourcifAI addresses the growing demand for transparent, data-driven decision support in the context of sustainable development and environmental resource management. Climate change and environmental degradation increasingly challenge decision-makers to interpret large, heterogeneous data sources and to derive robust, decision-relevant indicators for risks, impacts, and potential actions.
While transformer-based models offer strong capabilities for analyzing complex environmental data, their results often remain difficult to interpret and control in practice. Existing approaches typically lack user-centered, interactive mechanisms that allow decision-makers to explore model outputs, understand their implications, and actively influence the analytical process.
ResourcifAI develops a Visual Analytics–based decision-support system that directly couples transformer-based models with interactive visualizations. The system enables users to explore extracted environmental indicators, interactively steer the underlying models, and immediately observe the effects of their interactions in the visual interface. By allowing adjustments to inputs, parameters, or analytical focus, the system supports iterative sensemaking, improves transparency, and fosters trust in AI-supported analyses.
The project follows three closely connected research directions:
Optimization of transformer-based models for the extraction of reliable and context-sensitive environmental indicators from heterogeneous resource data.
Design of an interactive Visual Analytics approach that enables real-time, user-guided steering and interpretation of model behavior.
Extension towards comparative analysis, allowing the assessment of alternative actions and scenarios, including their potential impacts, uncertainties, and trade-offs.
Together, these components aim to support informed, sustainable decision-making in complex and uncertain environmental contexts.
Core Features
Transformer-based extraction of environmental and sustainability indicators from heterogeneous resource data
Interactive Visual Analytics interface with bidirectional coupling between analytical models and visualizations
Real-time user steering of model behavior through visual interaction (e.g., focus, parameters, inputs)
Visual exploration of impacts, uncertainties, and multi-dimensional indicator spaces
Comparative analysis of alternative actions and scenarios to support decision-making under uncertainty
Iterative, evaluation-driven prototyping, including empirical user studies on usability, transparency, and decision support
