Avery Hall Basement Rearrangement

A study of optimizing possible arrangements for review/exhibitions in Avery 100.

Team Members: Cohaul Chen, Walter Wang, Wenjing Tu, Elena Yu, Zixiao Zhu

1. Abstract

Avery Hall was built in 1912 as part of the original campus plan. As the architecture department has expanded, the available space for conducting studio reviews has become increasingly cramped. This lack of space has forced students and faculty members to share classrooms and other inadequate spaces, which can negatively impact the quality of reviews and limit the department’s ability to foster collaboration and innovation. This is a pressing issue that needs to be addressed to ensure that the department can continue to provide a high-quality education for its students and maintain its reputation for excellence in the field of architecture. This project aims to highlight the underutilization of the Avery basement and propose potential solutions to optimize its use for studio review or exhibition. Therefore, the project uses generative design as a method to generate optimal layouts for reviews.

2. Introduction

Before proposing any solutions, it is important to assess the current condition of the basement. The existing space is limited and has been subject to various constraints. (Fig.1)For instance, the Wallabout Cafe is located nearby, which may cause disturbance to people who are conducting reviews. The staircase to the wood auditorium obstructs the flow of traffic and makes it inconvenient for people to stay on. After conducting a thorough analysis of the basement’s strengths and weaknesses, we have identified specific walls that offer the most potential for improving the space for review purposes.

Fig. 1 Usable Wall

The project user is the administration of GSAPP, who is responsible for managing the resources and facilities of the school. By applying an automated system, it would be easier for them to identify the most efficient and effective way to organize reviews and exhibitions for multiple studios.

3. Methodology

Based on the idea, we propose a four-step process:

1. Collect input information and calculating maximum wall space for each student
2. Determine class boundary based on proximity to the walls.
3. Apply view angle and visibility as metrics to measure the performance.
4. Generate the optimal options and provide users with options.

Fig. 2 Mindmap

3.1 Input Parameters

To determine the maximum wall space available per student,

1. We first calculated the total number of students by multiplying the Number of Classes and Student Number per Class inputs.
2. Then we divide this value by the total length of the wall.
3. With the maximum wall width per student determined, we divided the wall into segments.

Fig. 3 Wall Division

3.2 Poster Clustering and Class Boundary

To create a comfortable distance between the jury and the wall during reviews, we created a boundary by offsetting the wall by 7 feet. Within this narrowed down boundary, seeds were randomly populated, from which a given number of seeds were selected as the center point of a cluster of chairs for the juries and students. The number of points is determined by the Number of Class input. This approach ensures a fair and equal distribution of seating for each class during reviews.

Fig. 4 Poster Clustering and Class Boundary

With walls divided and cluster points selected, we measured the distance between the middle point of each wall segment and each of the cluster points, and connected the middle points to their closest cluster points.

Fig. 5 Before and After

Fig. 6 Two Classes Layout

Fig. 7 Three Classes Layout

3.3 Minimum number, View angle and visibility

First, We calculated the angle between the visual line and normal line as θ. A smaller value of θ indicates better visibility for the jury. We added all the θ values. The optimal option was then identified as the configuration with the smallest sum of θ, ensuring that all juries have the best possible visibility during review/exhibition. For each iteration, the sum of θ is calculated towards the score for optimization.

Fig. 8 Normal Line (blue) and θ

Second, the total number of lines in each cluster is compared to the desired number of students in each class. When a cluster with lines fewer than that number, a penalty number will be calculated towards the score for optimization in this iteration.

Third, when the lines in a cluster intersect with columns, the intersections are documented and counted. The counts of intersections are weighted and calculated towards the score for optimization.

Fig. 9 Minimum Number & Intersection of Curves and Circles

3.4 Optimization

As mentioned earlier, view angles, minimum number, and visibility are factored in when calculating a final score for each iteration. In this project, the minimum number of lines has the heaviest weight as meeting the number of students per class takes priority. In the end, a higher score indicated higher penalty, which means a less optimal layout for reviews.

Fig. 10 Optimization

4. Conclusion

Our automated model, which utilized Rhino, Grasshopper, Python, and Galapagos, proved to be efficient and effective in optimizing the review/exhibition space in Avery basement by incorporating assessment of spatial proximity, view angle, and visibility. By collecting input information, generating wall divisions and layouts, and dividing classes, we were able to create an optimal arrangement that maximizes space and visibility for all students and faculty. More importantly, our model can also be applied to other fields, such as art gallery layout with given available wall spaces and number of viewers. The use of computational tools offers architects and designers a valuable means to optimize the use of available space and improve the user experience. Our approach demonstrates the potential of computational design methods to solve complex spatial problems in innovative and efficient ways.