Evaluator Node 1: This metric assigns a weighted value to floors based on their zone (first 25% of floors, middle 50% of floors, top 25% of floors). This logic posits that the top 25% of floors are the most valuable floors for real estate due to their better views, exclusivity, and reduced street-level noise, so, I weighted the value of the floors in this area by multiplying them by 1.5. I assigned the bottom 25% of floors a weighted value by multiplying them by 1.3, as these floors provide retail opportunities. I left the middle 50% of floors unweighted.
The rest of the node calculates the floor area within each zone, applies the appropriate weight, and sums the weighted values to compute a single Zoned Floor Value for the building. This final value can be used as a metric of desirability on the real estate market. This node also automatically adjusts the number of floors in each zone based on the total number of floors in the form, ensuring the logic scales regardless of building height, as shown in the final output which tests a number of different building heights.
Evaluator Node 2: This metric measures how volumetrically compact the building is by comparing the building’s volume to the volume of a bounding box with the same footprint. This node assumes that an extruded rectangle is the most compact and therefor the most economic and easy to build shape, and how much the volume of the actual form deviates from its bounding box is an indication of how much more inefficient it will be in terms of structure, material use, and envelope surface area, which are all important considerations for construction cost and energy performance. The node begins by extracting the actual solid geometry of the mass and calculating its true volume. It then computes the volume of the bounding box by identifying the minimum and maximum corner points and multiplying the box’s width, depth, and height. It then divides the form’s actual volume by the bounding box’s volume to produce a volumetric compactness ratio. A value closer to 1 indicates a more compact, efficient form, and lower values suggest a more inefficient geometry.
Points to ponder: I think the new evaluation metrics I designed do show meaningful differences between the building form alternatives, especially when looking at the results in the Excel table below. For example, some test cases that had a high Gross Floor Area didn’t score well in terms of Zoned Floor Value, since a lot of their area was in the less valuable middle zones. Similarly, forms with more extreme rotations tended to have lower Volumetric Compactness, even if their total volume was similar. These metrics also showed that the top-performing designs weren't just the tallest or largest volume. Instead, they were the ones that balanced between desirable floor area and volumetrics efficiency. That distinction would’ve been hard to make without my custom metrics. That being said, adding other measures like solar gain or construction cost would also help create a fuller picture of performance, especially in a real-world decision-making context.
For my single-objective optimization scheme, I combined these metrics: Gross Floor Area (GFA), Zoned Floor Value (ZFV), and Volumetric Compactness (VC). I normalized these metrics using min-max scaling, and weighted each based on its relevance to the project’s goals (multiplying ZFV by 0.5, GFA by 0.3, and VC by 0.2.) I weighted Zoned Floor Value the most at 0.5 because it reflects which parts of the building will be most valuable to a developer from a real estate standpoint. I gave Gross Floor Area a 0.3 weight since we still need to hit the target square footage, and weighted Volumetric Compactness at 0.2 because it’s useful for understanding efficiency, but not as important as meeting the space and revenue goals.
The top design alternative was Height = 800 ft, Rotation = 15deg, and had the highest composite score of 0.994. It performed best in terms of maximizing desirable floor value and spatial efficiency. Alternatives with high rotation performed worse due to lower compactness and more costly upper-level floors. While this is an interesting result, it fails to take into account any of the aesthetic choices that might cause a designer or developer to favor a more sculptural design.
Points to Ponder:
- My method was to normalize each of the evaluation metrics and then assign weighted values to each based on how important they were in relation to the project's goals (creating valuable square footage, staying within a desirable floor area range, and avoiding inefficient forms that will increase construction and operating costs.) By normalizing the metrics and assigning weights based on those priorities, I was able to combine them into a single score that made it easier to compare test cases at a glance, which I believe easily and helpfully captures the relationship between the most prominent evaluation metrics,
- I chose the top alternative because it offers the best mix of high-value floor area and volumetric efficiency, without sacrificing too much in other areas. It isn’t the tallest or the biggest, but it balances all of the important metrics I mentioned above. That being said, quite a few more metrics should be measured before committing to this form, especially for a site in Dubai, where building form, energy efficiency, cooling abilities, views, etc. can have a big impact on a building’s overall performance.