Step 1 - Generative Design Framework
The structure that I am evaluating is the bus shelter from module 3. Just like with any structure there are design options that can be evaluated:
- Structural Performance
- Design Variables
- Design loads
- Sizing of members
- Capacity of members
- Evaluators
- Demand capacity ratios can be calculated to determine the efficiency of members. By pushing these ratios closer to 1 the member sizes can be decreased creating an efficient design.
- Plastics hinges can be implemented which allow the structure to behave nonlinearly. This means that the member capacities can be decreased and therefore their sizes as well.
- The cost it takes to produce the member sizes also factors into the structural performance.
- Most Important Tradeoffs to Consider
- • The most important tradeoff to consider is the relationship between loads, costs, and member sizes. This is because the larger the member sizes the larger the cost is going to be. However, by decreasing some member sizes it decreases the material cost but could result in increasing other member sizes. Basically, the goal is to decrease just enough member sizes so that the cost decreases, but the structure can still withstand the demands.
- Architectural Design
- Design Variables
- Amount of Shading (Overhang Length and Height)
- Protection from Weather (Overhang Dimensions)
- Structure Orientation
- Material Choice
- Evaluators
- Number of direct hours of sunlight
- Wind speed experienced by the structure
- Visibility
- Most Important Tradeoffs to Consider
- The most important tradeoff to consider is the relationship between protection from weather (such as shading, wind, sunlight, etc.) and material choice. This is because the more protection from weather that is desired the more material that would be necessary. Additionally, this could lead to a less appealing structure.
- Sustainability
- Design Variables
- Material Choice
- Material Usage
- Material Cost
- Operating/Maintenance Cost
- Evaluators
- Carbon footprint
- Carbon emissions
- Greenhouse gas emissions
- Most Important Tradeoffs to Consider
- The most important tradeoff to consider is the relationship between material choice/cost and carbon footprint. This is because a lower carbon footprint might result in a higher cost for materials and their usage. It will also have an effect on the certain materials that can be used.
Step 2 - Generative Design Study
The design decision that I chose was trying to maximize the surface area of the roof while trying to minimize the maximum wind speed that the structure would experience. These design decisions relate to the architectural design and the structural design framework. Specifically, they focus on material cost, member sizes, and protection from weather. The purpose of this design was to provide enough shelter from the weather without increasing the cost too much or negatively affecting the structural integrity.
As stated earlier the shelter from Module 3 was used with a slight change. The main difference is that the columns and beams along the side have been removed. The images below display the code that was used during the generative design process.
In order to do this generative design study, the Galapagos tool was implemented. By modifying genome nodes the block can find the optimal outcome based on the desired fitness function. The genomes for the code above were structure height and structure depth. This is because structure height played a role in both the surface area of the roof and the wind speed calculations. It is important to note that in order to do the wind speed analysis a structure height less than 10 meters cannot be used. This is because the EPW file records wind speeds at a structure height of 10 meters. The fitness function for the code is the subtraction of wind speeds from structure height. Due to the difference in units between the two variables a factor could be applied to find a better generative design.
Step 3 - Generative Design Study Results
The optimized shelter is shown below:
The below image is of the Galapagos tool design process:
The Galapagos tool was iterated over 41 generations and reached an optimal design around iteration 10. In theory, the tool could be run until all possible iterations were completed. However, since the inputs are integers there are infinite possibilities. Therefore, the tool was terminated once there was little to no change in the designs. This corresponds to a flat line on the top graph in the image above. The peak value on the top graph from above or the top green bar on the bottom right from the image above indicate the optimal design. The line graph in the center indicate the values over time and displaces their magnitude on the image to the right. The scatterplot on the left of the image indicates possible design options (similarity representations). If there is an “x” that means that iteration was not able to produce a design after that because it would not be an optimal design. Whereas the blue dots mean that the iteration was able to produce an optimal design. If the markers are close together, it means that the optimal design converges around similar values.