Rizka Oktaviani

For this module, I first considered three possible design decisions that can be studied through a generative design framework. I wanted to focus on design problems that are simple enough to model, but still show clear tradeoffs between cost, energy, carbon, and performance.

• Design Decision 1: Solar PV roof design for a simple building
• Design Variables
• Annual energy demand
• Cost of grid electricity
• Panel efficiency
• Number of floors
• Usable roof percentage for PV
• Evaluators
• Payback period
• Annual solar energy coverage
• Total system and electricity cost
• Carbon emissions
• Most Important Tradeoffs to Consider
• More usable roof area and higher panel efficiency can increase solar energy coverage, but may also increase system cost.
• A larger PV system can reduce carbon emissions, but the payback period may not always be the shortest.
• More floors increase total energy demand, which makes it harder for the roof PV system to cover a high percentage of the annual energy use.
• Design Decision 2: Building massing and envelope efficiency
• Design Variables
• Building width
• Building length
• Building height or number of floors
• Roof form
• Evaluators
• Surface area
• Volume
• Envelope efficiency
• Energy demand proxy
• Most Important Tradeoffs to Consider
• A larger building can provide more floor area, but it also increases envelope area and energy demand.
• A more compact form may reduce heat transfer, but it may not always provide the desired architectural form or roof area for PV.
• Taller buildings may use land more efficiently, but roof area becomes smaller relative to total energy demand.
• Design Decision 3: Façade glazing and shading strategy
• Design Variables
• Window-to-wall ratio
• Shading depth
• Glass performance
• Orientation
• Evaluators
• Daylight/view score
• Solar heat gain proxy
• Façade cost
• Cooling load proxy
• Most Important Tradeoffs to Consider
• More glazing can improve daylight and views, but it can also increase cooling load and façade cost.
• More shading can reduce solar heat gain, but it adds material, cost, and may reduce daylight.
• The best option is not necessarily the most glass or the deepest shading, but the balance between comfort, cost, and energy performance.

For the generative design study, I chose to study the solar PV roof design because it connects directly to energy, cost, and carbon performance. The model is intentionally simple, but it still shows a useful design tradeoff. The building geometry is based on a rectangular building form, where the number of floors and roof area affect how much energy the building uses and how much PV can fit on the roof.

The main design variables I used were annual energy demand, cost of grid electricity, panel efficiency, number of floors, and usable roof percentage for PV. I also used constants such as building square footage, grid carbon intensity, inverter efficiency, panel cost, and panel lifespan. The Dynamo graph calculates the available roof area for PV, the number of panels, the potential annual solar generation, the percentage of annual building energy covered by solar, the total cost, the payback period, and the resulting carbon emissions.

The objective of the study is to understand the tradeoff between maximizing annual solar energy coverage and minimizing payback period, total cost, and carbon emissions. This is important because the “best” PV design is not simply the largest system. A larger system may produce more solar energy, but it may also cost more or take longer to pay back. The study helps compare which combinations of roof coverage, panel efficiency, energy demand, and electricity price create the most balanced design.

The scatterplot shows the relationship between Payback Period on the X-axis and Annual Solar Energy Coverage on the Y-axis. I used Carbon Emissions as the bubble size and Total System and Electricity Cost as the color. This makes the tradeoff easier to read because I can see not only which options have better solar coverage or shorter payback, but also which ones may lead to higher cost or higher emissions.

From the results, the options with shorter payback periods are not always the same options with the highest solar energy coverage. Some alternatives produce more solar energy, but they also have higher total cost or larger carbon-related impacts because the overall building energy demand is higher. In general, I would look for options toward the left side of the scatterplot with relatively higher solar coverage, smaller bubble size, and lower-cost color. These options would give a better balance between financial payback, energy performance, and carbon reduction.

The parallel coordinates graph also helps show how each variable affects the outputs. For example, higher panel efficiency and higher usable roof percentage generally improve solar energy coverage, but the number of floors and annual energy demand strongly affect whether the roof PV system can cover a meaningful percentage of the building’s load. I would use this information to avoid over-sizing the building energy demand without also increasing the PV potential, and to focus on roof efficiency and panel efficiency as the more useful levers for improving performance.

The Dynamo study graph was organized into inputs, building geometry, energy use, solar PV design, and outputs. This helped me keep the model logic clear: the input variables drive the building size and PV assumptions, then the formulas calculate the final evaluators used in the generative design study.