“ArchiGenOpt: a Parametric Design & Generative Optimization Tool for Building Design”
Tool Overview
The proposed tool is designed to help users both create building designs and select the best design by leveraging parametric design and generative optimization techniques using Dynamo. It enables users to model a simulation environment that represents the project site context, create building designs, and explore a wide range of design options based on footprint shape, dimensions, height, position, orientation, and facade material. By integrating multiple performance criteria, the tool facilitates informed decision-making and promotes efficient and sustainable design practices. Intended users include architects, VDC managers, urban planners, architectural engineers, design consultants, real estate developers, and clients.
Intended users
- Architects & Building Designers
- VDC Managers
- Urban Planners
- Architectural Engineers
- Design Consultants
- Real Estate Developers
- Clients
Need you’re Trying to Provide a Solution or Support for
When embarking on building projects, design professionals face the challenge of balancing multiple conflicting objectives such as cost, aesthetics, functionality, sustainability, while also having to comply to site limitations and design constraints (like the maximum allowed gross floor area of the building). These challenges are further complicated by the necessity to incorporate various datasets, such as geographic information, material properties, energy consumption data, environmental impact assessments, and cost assessments. Moreover, they must manage numerous design iterations and facilitate effective communication between stakeholders. The proposed tool aims to streamline this process during the early design phase by providing an integrated platform that optimizes building designs based on multiple performance criteria, facilitating more informed decision-making, and promoting efficient and sustainable design practices.
Contextual Setup
Before defining the design variables (inputs), the user will be able to create a simulation environment, representative of the project site context. Specifically, the user will be able to:
- Model the surrounding environment where the proposed design will be built, by:
- either uploading a Revit link (importing a detailed model of the site and surrounding area),
- or adding obstacle buildings directly in Dynamo. The user will be able to specify the number of buildings to create around the site and define their positions and dimensions.
- Define site constraints:
- the user will be able to define the site constraints, i.e., the boundaries and limitations of where the proposed building can be placed. This will be defined as a rectangle by specifying the length and width of the site area.
- Define Viewpoints: Viewpoints are specific locations around the building site used to analyze the visual impact and sightlines of the proposed design. They help ensure that key views, such as landscapes or oceans, are preserved. By defining viewpoints, users can simulate and evaluate the visual experience from various positions and heights around the building. Each viewpoint in Dynamo is modeled as a series of points distributed along a vertical line aligned with the Z-axis. From each of these points, a certain number of rays depart radially and parallel to the xy plane. The user will be able to define the viewpoints by specifying:
- The number of viewpoints: the user will be able to add up to 4 viewpoints, one for each side of the compass (north, south, east, and west), and place them around the building.
- Each viewpoint’s parameters:
- (x, y) coordinates: the position of the viewpoint.
- Max height: the maximum height reached by the last point of the vertical series. This point will have the same height of the maximum allowed building height, also specified by the user.
- Number of points: the number of points placed on the vertical line from 0 to the max height.
An example environment that can be created by the user is illustrated below:
Inputs
Once the user has defined the environment, they can specify the following design variables (inputs):
- Variable input 1: building footprint geometry - Rectangular, polygonal, cylindrical, irregular (predefined shapes).
- Variable input 2: building footprint dimensions (feet) - Assuming regular shapes, either measure length and width for rectangles and polygons, the radius for cylindrical buildings, predefined parameters for irregular shapes.
- Variable input 3: building height (feet)
- Variable input 4: building orientation (degrees)
- Variable input 5: building position (x, y coordinates in feet) - Displacement from a defined origin point along the x and y axes, measured in feet.
- Variable input 6: material type - Select from a predefined list of materials (e.g., reinforced concrete, steel, timber). Note that each material will have an associated emission factor for calculating embodied carbon (see Outputs section).
Underlying Logic of the Model I Will Implement
The model operates on a framework of parametric design and generative optimization, leveraging computational algorithms to iteratively explore a vast design space while adhering to specified constraints and objectives. Here is an overview of the key components and their interactions within the model, in order of workflow:
- The user starts by defining the project site and its surroundings, either by importing a detailed model or by specifying obstacle buildings directly within the tool.
- Site constraints, such as boundaries and limitations, are then defined to establish the spatial context for the building design.
- Viewpoints are additionally established to analyze the visual impact and sightlines of the proposed design from various perspectives (maximum four viewpoints allowed).
- Users input various design variables, including building footprint geometry, dimensions, height, orientation, and position.
- These variables define the desired design parameters and also serve as the initial conditions for the generative optimization process.
- Dynamo offers a “Generative design” tool that systematically generates and evaluates diverse building designs by dynamically adjusting input variables.
- This tool is equipped with generative optimization algorithms to assess design alternatives against predefined objectives and constraints, ensuring alignment with project goals. The objectives can be defined by the user by maximizing, minimizing, or by ignoring the outputs that are defined in the next section.
- Filters are finally applied to visualization outputs, such as parallel coordinate graphs, to align with optimization objectives, aiding in interpreting and comparing design variations, facilitating informed decision-making, and final design selection.
Outputs (& Suggested Optimization Objectives)
- Output 1: Total Construction Cost
- The total construction cost represents the sum of construction costs for all floors, considering the cost per square foot linearly increasing with increasing elevation.
- Note: for this initial implementation of the tool, assuming a linear relationship between construction costs and elevation is acceptable. While this simplification may not capture all nuances, it provides a reasonable starting point for cost estimations. Subsequent iterations of the tool can incorporate more sophisticated cost models as needed.
- Output 2: Solar Efficiency Index
- The solar efficiency index metric evaluates the building's ability to harness solar energy efficiently throughout the year, providing insight into the building’s potential for passive solar heating, daylighting, and photovoltaic energy generation. This index is calculated by measuring the cumulative solar insolation available on the building envelope surfaces and comparing (dividing) it to the total surface area of the envelope.
- Output 3: Number of Sightlines (Rays)
- This output measures the number of sightlines (rays) intersecting the building from designated viewpoint(s), accounting for fixed surrounding structures that may obstruct views.
- Output 4: Embodied Carbon Estimate
- This metric provides an estimate of the total embodied carbon of the building envelope, based on the selected material. The calculation involves determining the volume of the solid building envelope and applying an emission factor specific to the chosen material.