Stage 1 – New Evaluator Metrics
I created two evaluator nodes to assess the building form alternatives generated in Module 5. The two metrics selected were form factor and incident solar radiation on the building envelope, including the façade surfaces and the roof.
The first metric, form factor, was calculated as the ratio between the gross envelope surface area and the gross floor area:
Form Factor = Gross Envelope Surface Area \ Gross Floor Area
This metric was selected because it indicates the compactness and geometric efficiency of the building form. A lower form factor means that the building has less external envelope area per unit of usable floor area. This is desirable because the envelope is the interface through which heat transfer occurs. In Dubai’s cooling-dominated climate, reducing envelope area can help reduce heat gains and improve the environmental efficiency of the building.
The second metric, incident solar radiation on the envelope, was calculated using the Ladybug Incident Radiation component. This evaluator measured the cumulative solar radiation falling on the façade and roof. This metric was selected because solar exposure is a major contributor to cooling demand in hot climates. While form factor evaluates the compactness of the geometry, incident radiation evaluates how the geometry interacts with solar exposure, including the effects of orientation, twist angle and self-shading.
Together, these two metrics provide complementary information about environmental performance.
To implement these evaluators, custom Grasshopper nodes were created and integrated into the Anemone testing loop developed in Module 5. For each iteration, the workflow generated the building geometry, calculated the evaluation metrics, and stored the results in a summary table together with the input parameter values. The reported outputs included:
- Input parameter 1: building height
- Input parameter 2: building twist rotation angle
- Gross Floor Area
- Gross Surface Area
- Form Factor
- Incident Solar Radiation
Stage 2 – Single-Objective Optimisation Scheme
For the single-objective optimisation stage, I developed a weighted scoring system to combine the different evaluation metrics into a single value. The objective was to identify the building form that provided the best overall balance between development value and environmental performance.
The three main objectives considered were:
- Maximise gross floor area
- Minimise form factor
- Minimise incident solar radiation on the envelope
These objectives are partially aligned. Increasing building height increases gross floor area, which improves the development potential of the project. However, increasing the building size also increases envelope area and solar exposure. Similarly, twisting the building increases the surface area of the envelope. As a result, the optimisation problem involved balancing competing objectives rather than maximising a single parameter.
Because the metrics used different units and numerical ranges, they could not be combined directly. Gross floor area was measured in square metres, form factor was a dimensionless ratio, and solar radiation was measured in kWh. To address this issue, all metrics were normalised using min-max normalisation, converting the values into a common 0–1 scale:
X_norm = ( X - X_min ) / ( X_max -X_min )
For gross floor area, higher values were considered better, so the normalised values were used directly. For form factor and incident radiation, lower values are preferable, so the normalised values were inverted:
FF_score = 1 - (( FF - F_min ) / ( F_max - F_min ))
SR_score = 1- (( SR - SR_min ) / ( SR_max - SR_min ))
An important challenge encountered during this process was that the normalisation required knowledge of the minimum and maximum values for each metric for all design alternatives. These values were not known before the simulations were run. As a result, the workflow had to be divided into two stages.
First, the Anemone loop was used to generate and evaluate all building alternatives while storing the values for gross floor area, form factor, and incident solar radiation. After all simulations had been completed, the minimum and maximum values for each metric were identified. These values were then used to normalise the metrics and compute the final weighted score for every alternative.
The final combined evaluation score was calculated as:
Score = 0.4 * FAS + 0.3 * FFS + 0.3 * SRS
where:
FAS = Floor Area Score FFS = Form Factor Score SRS = Solar Radiation Score
Gross floor area was given the highest weighting because it represents the economic value of the project. Form factor and incident radiation were also given significant weighting because they relate to environmental performance and potential cooling load. The weighting scheme aims to balance development efficiency with environmental considerations.
The table shows the parameters evaluated and the top 3 performers (best in green, second and third best in yellow)
