For this module, the original building (Figure 1) was optimized according to a given set of evaluation metrics, which resulted in a finalized optimized building (Figure 2). The optimization process involved a parametric study, begun in Module 5 that looked at the varying effects of building height as well as varying sizing of the top level dimensions. The forms produced from this were then evaluated based on the expected cost and revenue of the project over a 50-year life span. Ultimately leading to the optimized solution shown below.

When breaking down costs and profits for any facility, three primary driving factors include the initial construction costs, profit from rent, and operational costs.

(1) Construction Costs: this will look at the cost to construct our form where pricing will increase linearly over the height of the building.

(2) Profit from Rent: this will look at the average cost of rent/sf for commercial space in the southern market district (which falls around $75/sf). This average was used as a baseline for rents for my proposed building.

(3) Solar Insolation Potential: this will look at the average solar insolation of the building, which will reflect the potential of the roof to harvest energy, which would help to decrease operational costs.

In order to understand the financial feasibility of a project it is crucial that we understand how these factors break down in order for us to determine the optimal building to construct.

Starting with the overall logic of the Dynamo script we see that the script is very similar to that developed in Module 5. The script begins in the green section by taking in input values for the parametric study discussed in Module 5 Part 2. In addition to these inputs, new inputs for evaluation of construction costs, rent profits, and solar insolation have been included. From this section, the combination of parameters to be studied are created in the orange portion. Then, these values are evaluated based the aforementioned evaluation parameters in the light purple custom node. Each of these evaluation functions are tested, and then the desired results are then selected to be displayed and printed into excel.

A closer look at the input parameters for the evaluation methods:

At the start we have the input values associated with each of the evaluation metrics construction costs (Figure 4), rent profits (Figure 5), and solar insolation (Figure 6).

The construction costs will be evaluated based on a linear interpolation between ground level costs of $700/sqft and 750ft elevation costs of $1500/sqft.

The rent profits were determined by splitting the building into three sections. A bottom portion between 0 - 200 ft with rental costs linearly ranging from $50-60/sqft, a middle segment between 200 - 500 ft (where the average SF skyline falls) with rental costs linearly ranging from $60-75/sqft, and finally the premium office space located on average above the skyline providing clear views from 500 - 750 ft with rental costs linearly ranging from $75-90/sqft.

The solar insolation input values include the weather based on the buildings San Francisco site location, the spring sun setting conditions chosen in Revit, critical shading surfaces from the surrounding buildings, and a selected spacing between points of solar analysis. These values will help to determine the solar insolation of our building form.

A closer look at the costume node evaluation parameters:

The custom node created takes in the inputs values described above and from Module 5 to test various building forms based on input heights and top side lengths, and evaluate them according to construction costs, rent, and solar insolation.

From Module 5, we created the dynamo logic that would create the updated building form based on the given parameter test case instances (seen in the orange and light purple sections on the top of Figure 8). The critical values of gross floor area, surface area, and volume were reported in Module 5 (light pink), but in this module we have the addition of our previously described evaluation metrics. These additions will be discussed in further detail below.

For the first evaluation, we will take a look at construction costs (Figure 9). The evaluation process was placed into a costume node seen in Figure 10, which uses a linear interpolation method to determine a multiplier that would then be applied to the associated mass floor area. This determined the cost of construction level by level. Any null values were removed, and the remainder were then summed together to find the overall cost of construction for each building form.

For the evaluation of the rent profits (Figure 11), we have taken the overall buildings and divide it into three segments with varying ranges for rent. The logic for this process was encapsulated in a custom node seen through Figure 12 and 13. This process was similar to the construction costs, in which the rent amounts were linearly interpolated to create a rent multiplier. The values determined from each of the three segments was then combined and applied to their associated mass floor areas. Any unused null floor mass values were removed and the reminder were then summed together to find the rent for the building for a given month.

For the evaluation of solar insolation, I utilized the weather and sun path data from the Revit environment to compute the cumulative solar insolation of the building based on the building’s roof surface as this surface would be the location of potential solar panel placement. The logic for this evaluation process, shown in Figure 14, firstly selects the roof surfaces and passes this roof along with the solar setting and area of the roof into a custom node. Figure 15 and 16 show us a closer look into the custom node. I modified this node to return the amount of money that would be saved by the solar potential of the roof. Figure 16 shows the logic of taking the amount of cumulative potential (kWh/m^2) and multiplying that by the area of the roof (ft^2) and the cost per kWh ($/kWh) to determine the amount of potential money saved. The roof potential was analyzed based on the shading created by the surrounding buildings as shown in Figure 17.

Once the dynamo script initial evaluation process is completed, the results for each parameter pairing is exported to excel. The analysis that was performed tested building heights ranging from 550 ft to 650 ft with a top side length ranging from 80 ft to 90 ft. The initial results are summarized in the table below in Table 1.

With these values computed, the next step was to create a weighted single-objective optimization scheme. In this module, we have the added evaluations of construction costs, monthly rent income, and average solar insolation potential. In order to properly compare these values and produce a singular best case, some manipulation of the result data is required.

To start off, the values computed for this module were initially selected as a means to compare the buildings upfront construction costs with the continuing rent profits and operation costs associated through a 50-year life span of the building. These three values will first be normalized according to the maximum value found for each case, and then multiplied by a assigned multiplier that is weighted more heavily on the recurring rent and solar insolation potential in comparison to a singular construction cost, as these metrics relate to the profit/performance of the building over its entire life. Between rents and solar potential savings, rent will have a larger weight as it will more significantly impact the profits of the building. (The scaling values used were construction cost 20%, rent profits 50%, and solar potential savings 30%). The scaling process is shown in Table 2 and the results are shown in Table 3.

Based on the selected parameter choices and single-objective optimization scheme it turned out that the optimal solution option improved through the iterations of parameters. Therefore, the optimal solution would be a building height of 650 ft with a top side length of 90 ft. Reflecting on these results, in order to have more interesting trade-offs, we could explore different solar studies that could change the optimization scheme.