Description

Problem

Simulation-based Digital Twins are increasingly used to support monitoring, prediction, and decision-making in complex systems such as manufacturing and energy infrastructures. While simulation models are traditionally considered transparent, modern Digital Twins combine multiple data sources, complex models, and automated model extraction pipelines, which can make their behavior difficult for users to understand.

Explainability has been widely studied in artificial intelligence, but its application in Digital Twins remains fragmented. Explanations in Digital Twins may take many forms, including textual descriptions, visual dashboards, immersive environments, interpretable models, or multimodal interfaces.

Goal

The objective of this thesis is to conduct a PRISMA-compliant systematic literature review on explainability in simulation-based Digital Twins. The study will analyze how explanations are provided to users across different modalities, levels of abstraction, and components of the Digital Twin Framework. The review will identify key approaches, technologies, and user groups involved in explainable Digital Twins, will highlight research gaps and opportunities for future work in this area, and produce a framework and recommendations as the outcome.

Research Questions

1. How is explainability in simulation-based Digital Twins operationalized in the literature with respect to the object of explanation, explanation delivery mechanism, user interaction mode, timing of explanation provision, and level of abstraction?

2. How are explainability outputs distributed across the DT component and across relevant stakeholder groups, and what principal gaps can be identified?

3. What design recommendations emerge from the reviewed literature for developing explainable simulation-based DTs, and what directions should future research prioritize?

Required Skills and Knowledge

• Research and analytical skills for conducting systematic literature reviews
• Ability to review and synthesize technical literature

• Interest in Digital Twins, simulation modeling, and human–system interaction.

Depending on the scope and available resources, the thesis may also include stakeholder surveys or interviews to identify explainability needs, or a small implementation demonstrating explainability mechanisms within a Digital Twin environment.