On account of the 2nd prize of VREVAL in this year's AVRiL competition, Lila Becker interviewed Sven Schneider on the use of the the VR-based user study tool VREVAL for the education of future architects.
Please find the full interview here:
In the following, you find an automated translation of the german interview:
Prof. Sven Schneider and his team from the Bauhaus University Weimar have set themselves the goal of concretizing the "fuzzy" human-building relationship that is at the heart of every building design, and in doing so have placed future building users at the center. Because in the end, the building has to please the people who use it, not the planning team. Thanks to VR, future users are not only invited to take a virtual tour of the building, but can also provide direct feedback in the same step. The tool has been christened VREVAL. VR for Virtual Reality, EVAL for Evaluation (survey). We talked to Prof. Schneider.
Mr. Schneider, how would you describe VREVAL to a person outside the field?
In architectural education, the relationship between people and space is a very central topic. The architect must be able to put himself in the position of the future building users and know which architectural means he can use to meet the user requirements. However, it is often the case that the evaluation of such user-centered aspects is carried out rather intuitively. If the intuition then does not match the real user behavior, this leads in the worst case to non-functioning and unpleasant buildings. This was ultimately our motivation for our project VREVAL, namely to create a tool that allows the user perspective to be examined systematically - evidence-based, so to speak - directly in the design process.
VREVAL is a Virtual Reality (VR) - based tool for the creation, execution and evaluation of user studies in the design phase of buildings. User studies are mainly known from product design or web design. There it works very well, because you can build a usable prototype relatively quickly. In architecture, however, the production of such a prototype for purely test purposes is simply impossible for reasons of time and cost. That's why VR technology is a very interesting tool here, because it gives you the opportunity to experience a building design in real size. Of course, it was already possible to view a building in VR before VREVAL. But what was missing until now were methods for architects to systematically investigate how future users would experience the building and how they would behave in it. And that is precisely what is now possible with VREVAL. After all, it's not about how you, as the planner, find the building, but about how everyone else finds it. In the end, it has to stand up to the users and not to architectural criticism.
To what extent does VREVAL make it easier for students to think about the relationship between people and buildings in a more needs-oriented way?
Students use the VREVAL tool to investigate certain user-centered aspects. Take orientation in the building, for example. To do this, they identify key locations in the building and create tasks that ask study participants to find those locations in the building. The paths that participants take are recorded and can then be analyzed. For example, it is possible to find out at which point the future users have problems finding their way.
Another example is the study of the spatial effect. For this purpose, the students develop a study in which different variants for the design of a space are compared with each other. Take, for example, a train station concourse in variants with different ceiling heights. The study participants can then look at the different variants and rate them on a scale of light/dark, pleasant/unpleasant, beautiful/ugly, etc.
After such a study, the results can then be visualized directly in the planning software (BIM). The often rather nebulously discussed user-centric aspects are thus presented in concrete figures and diagrams. And that, of course, helps in the decision-making process. For example, the easily calculable factors that are often at the center of planning, such as construction costs, can be compared with the study results. Perhaps it would cost 100,000 euros more to raise the ceiling of a train station concourse one meter, but the future users would find this space much more pleasant.
As I understand it, the innovative aspect of VREVAL is not the virtual tour, but the systematic determination of the behavior and emotions of future users. The collected data is thus displayed directly in the BIM software. Paths can be displayed in the floor plan or results of surveys can be read directly. How was user-centered building taught to students before VREVAL?
In classical architectural education, the student usually sits down with his or her design and discusses it with a lecturer. In the process, all possible aspects are discussed, from building construction and costs to aesthetics and user behavior. The latter two are often evaluated on the basis of the student's own experience and individual aesthetic preferences. A scientific basis is largely lacking here, which often leads to vague statements on the part of the instructors.
The interesting thing about VREVAL is that we as teachers actually stay completely out of the evaluation. So we don't give direct criticism, but the students have to work out this criticism for themselves. They have to build the tool that they use to examine how well their design works from a user's perspective, and then they have to analyze the results they get from that tool. In this way, students train their evaluation skills. The architectural education is sort of flipped. We don't tell students what's good or bad; they work it out for themselves. And the feedback doesn't come from a single person, but from the future building users.
So do you also see work being made easier in the design process?
Yes, definitely. The tool clearly makes it easier to visualize spatial dimensions and spatial relationships. Particularly for people whose spatial imagination is not as well developed. In this way, errors can be detected at an early stage, which also makes the entire subsequent planning process more efficient. But of course it does mean a certain amount of extra work in the early planning stages, because you have to take a close look at the perspective of the future user, and think about how to investigate certain aspects in a meaningful way.
Is it also possible to make direct changes to the environment in VREVAL? For example, by dragging a chair to where I would like to sit?
VREVAL is primarily about evaluating designs, not about designing itself. There are other, better tools for that. However, there are study modules in VREVAL in which participants are asked to place things in the room. For example, the question "Where would you like to sit?" is asked. The participants then mark these places with placeholder objects. The result is a map with the locations that the users prefer as places to stay. This information can then be used, for example, in the design of seating options. A study module is also being considered in which participants can make their own decisions about room design, e.g., by changing the size and location of rooms, windows, or uses. Here, however, the limited design competence of laypersons must be kept in mind: if they move a wall, for example, without considering the statics of the building, this is only of limited use.
Where do you see the greatest potential in the use of VR technology during a planning process?
The greatest potential probably lies in the communication of spatial environments. Hardly anyone can do anything with a 2D plan. There are even studies that show that even among architects there are major differences in the assessment of spatial qualities in different forms of representation. For example, floor plans and sections are often described in words that have nothing to do with the actual experience of space. In contrast, photorealistic renderings depict the designed spaces quite close to reality. However, the problem exists here that they always show only a certain section of the building. In addition, the perception of the actual room size is distorted in these images. In virtual reality, on the other hand, the rooms can be experienced in their true size and you can move around and look around freely. In this way, buildings can be examined and understood much more comprehensively by all those involved in the planning process.
How do lecturers' teaching approaches change as a result of using VREVAL?
As a development team, we've learned a lot over the last few years about the relationship between people and architecture and the failures in architectural education. If you're going to set up a user study with VREVAL, you first have to figure out what you want to find out in the first place. And that's where it starts. Many students often don't even know what qualities of use and experience they want to investigate. We have to talk to the students about this for a long time in order to sensitize them to the actual use of the building and to find out what they actually intend with their building design. Should it be about facilitating orientation, promoting social communication or creating a high quality of stay? Then you have to consider what architectural means have an impact on these qualities, and how to investigate whether these qualities are achieved with the design. This is often not addressed so specifically in traditional architectural education. So students realize that even though they've been studying for several years, they have their problems with that. But then after you go through this process once, it becomes much easier and they wonder why they didn't learn it much earlier.
To what extent is VREVAL currently being used at your institution?
We offer a seminar every semester and a so-called study project every year. In the seminar, students are introduced to the basic concepts. They then try these out on a small scale. Specifically, this is usually the redesign of their current home. So for example: What effect would a 50cm higher ceiling have? How would a wider hallway or an additional window feel?
In the study project, we then use VREVAL for a more complex design project. Through a long-standing and very productive cooperation with Deutsche Bahn Station & Service, we use real station designs as case studies here. The students then design a station building in the first half of the semester and spend the remaining time systematically evaluating this design. The goal is to gain insights from this evaluation to improve the design.
Are there plans for further development? If so, what do they look like?
The Thuringian Ministry of Economics, Science and Digital Society is currently funding our OpenVREVAL research project. The aim is to build a stable platform that can be used by everyone. The result will be a web-based tool that makes it easy to configure and coordinate user studies. A Google Forms for building models, so to speak. The project will run for another 18 months and then hopefully we'll be ready to make it public and anyone can use it. The challenge I see with this, however, is not so much technical, but lies in creating a meaningful user study. That's a big challenge for someone who has never done that before. So, in addition to the technology, what is needed above all is good training in user-centered design.