Master’s Thesis in MSc User Experience Design at the University of Applied Sciences of the Grisons

This master’s thesis investigates how Augmented Reality (AR) and automated 2D-to-3D workflows can be integrated into architectural planning processes in a user-centered manner and what added value they provide for clients and architects. The starting point is the recognition that traditional 2D plans frequently cause misunderstandings and late change requests, as clients’ spatial understanding is limited.

Expert interviews show that modern Computer-Aided Design (CAD) and Building Information Modeling (BIM) systems already enable rule-based, lossless model conversions, so that the essential added value lies in a media-break-free, user-oriented workflow. The direct transfer of CAD data into AR visualizations allows planning models to be experienced at 1:1 scale and in the real environment.

Methodological Approach
The study employs a Mixed Methods design with qualitative expert interviews, experience tests with an AR prototype for the Apple Vision Pro (AVP), and a quantitative online survey. The research design follows a sequential mixed-methods approach along a three-stage development cycle: exploration, prototype development, and evaluation.

In the first phase, semi-structured interviews with architects identified central requirements, needs, and pain points in everyday planning practice. The results served as the foundation for concept and prototype development, with an affinity diagram created to organize and synthesize the key findings. A multi-stage, semi-automated evaluation process was implemented for qualitative analysis to ensure efficient and comprehensible evaluation.

Persona and User Journey Development
Based on the expert interviews, two exemplary personas were developed for architects and two for clients. These summarize central characteristics, needs, and challenges of the relevant target groups.

The developed User Journey Map illustrates the most important phases, touchpoints, and potential pain points in the planning and visualization process from the user’s perspective. The journey clearly shows that particularly in the early and iterative phases, there are great potentials for immersive visualization solutions.

Prototype Development and Interaction Design
The AR prototype for the Apple Vision Pro enables clients to experience building designs in original size, location-based and immersively – both as exterior and interior experiences. The UI design is based on the official Human Interface Guidelines for visionOS and incorporates central principles of spatial, user-centered design.

The interaction concept is based on three native gesture controls as well as physical movement in the real environment. A central principle is the integration of real movement in space as the central navigation method.

To ensure physical safety, the floor area in the immediate vicinity is displayed transparently when walking movements are detected. This keeps real conditions such as unevenness, terrain steps, or obstacles visible at all times.

Additionally, a specific navigation function was implemented that combines drag gestures and plane detection. This allows the model to be explored efficiently even with limited physical mobility.

The interior view offers a multimodal navigation concept that flexibly adapts to different usage situations. The user can either move physically through the digital model or alternatively trigger virtual movement in space using hand gestures.

Evaluation through Experience Testing
The qualitative experience testing was conducted with five participants under practical conditions on a walkable property. The investigation used a multi-methodological design: Think-Aloud technique, semi-structured interviews, and standardized questionnaires including the NASA Task Load Index (NASA-TLX) for measuring subjective load and the User Experience Questionnaire – Short (UEQ-S) for measuring user experience and acceptance.

Quantitative Findings
For the quantitative survey, an online questionnaire was conducted with n = 44 participants. The survey used a within-subjects design with randomized order, where participants evaluated both classic 2D floor plans and AR visualizations on seven-point Likert scales.

The results show that AR visualizations significantly improve spatial understanding, decision confidence, client satisfaction, and communication between architects and clients compared to classic 2D floor plans. The AR visualization performed significantly better than the classic 2D floor plan in all surveyed aspects. Effect sizes range from d = 0.5 to d = 0.8 or higher.

Qualitative Insights
The qualitative evaluation shows that particularly the physical navigation and intuitive interaction in the AR model are positively evaluated. A central experience of all test participants was the extraordinarily high immersion achieved through the spatial experience of the building at 1:1 scale and in real context.

The evaluation of the NASA-TLX questionnaire reveals overall low to moderate subjective load when using the AR prototype. The evaluation of the UEQ-S shows that the AR prototype is predominantly positively assessed by participants. Particularly highlighted are aspects such as attractiveness, comprehensibility, and modernity.

Workflow Integration and Technical Feasibility
The research demonstrates the technical feasibility of an automated CAD-to-AR workflow. The conceived process integrates seamlessly into existing CAD environments through plugin architecture, automatically converting 2D floor plans and 3D models into AR-compatible formats. Both ArchiCAD and AutoCAD provide official APIs that enable native integration.
The workflow encompasses model export, format conversion to USDZ, cloud-based distribution, and direct loading into the AR application. This approach ensures a continuous user experience from modeling to immersive on-site presentation while maintaining the precision and accuracy of the original CAD data.
Acceptance depends significantly on simple integration into existing CAD workflows and the user-friendliness of the AR application. The study identifies practice-relevant requirements as well as limitations and provides recommendations for the further development of user-centered AR solutions in architectural planning.