For your final assignment in this class, you’ll design, develop, and share your own tool or application that showcases how parametric design, optimization, or generative design can be applied to an engineering problem of your choice.
The key idea is to develop your tool in such a way that you can share it with a community of colleagues who will be able to use or build upon your work for their own designs easily and efficiently.
What Kind of Tool or App Should I Create?
We’re using the term “tool” in a very broad sense. For our purposes, it can be anything that would be useful to another designer or user as they make design choices and decisions.
To be truly useful and convenient, you’ll want to create something that can stand alone. It should be:
Easily transferred – so other users can download it and apply it to their own projects.
Robust and reliable – it should work reliably within expected ranges of inputs and restrict users (or when them) to stay within the expected ranges.
Well-documented – you won’t be there to guide users through the specifics of using your tool. So, it should be well documented externally (for users who want to apply it without digging into the internal logic) as well as internally (for users who want to dive in and step through the nodes to explore ways they might adapt it to better fit their needs).
Some types of tools that you might consider developing include:
a ConfiguratorTool – that uses underlying rules to produce valid configurations for specified inputs. An example would be a tool that helps build up (or decompose) a design using standard sizes or configurations that can be easily manufactured off-site. So, you “configure” a design using available choices. Check out this page for another example of a Configurator Tool that configures Structural Frames for Variations of a Rectangular Building.
a Design Assistant – that provides useful feedback or recommendations that assist designers in evaluating the impact of potential options as they make design decisions. An example would be a tool that helps layout and place structural framing elements based on different inputs to guide the bay sizes or floor-to-floor heights. The tool could “assist” a designer by providing feedback about the weight of the structural materials or the carbon footprint for the inputs currently being considered.
a Generative Design Study – that evaluates a range of options for varying inputs and presents the results in a graphical way that helps designers identify the alternatives that best meet the desired criteria. You could expand on the study you created for Module 7 in many ways – adding inputs, defining new evaluation metrics, or refining the logic of your model – or create an entirely new study that looks at another design problem that you’d like to explore.
Unit difference:
As this is an open ended project/assignment, we are unable to prescribe exact guidelines for a different number of units, however we expect ~6h of work for 2 units ~9h for 3 units and 12h for 4 units. You are welcome to detail overall time spent in notion submission as well as proposed schedule/timeframe in the optional pitch proposal submission.
How Should I Get Started?
Here’s our recommended approach:
‣
Part 1 - Design Due Fri, May 29th [Optional]
You are welcome to submit this by Friday May 29th to receive feedback on your proposal ahead of working on the actual assignment submission. If you choose not to submit this by May 29th, it is then due with your final submission and feedback will not be provided other than in Office Hours should you ask for it.
‣
Step 1 - Choose an area of your interest…
Identify an area where you think a parametric design or optimization approach could be applied to create a useful tool to assist designers or decision-makers.
For example, you can focus on a tool to support:
Architecture / Engineering / Construction - related decisions
Sustainability or Environmental decisions
Or any area that’s of interest to you…
‣
Step 2 - Need-finding
Who’s going to use your tool?
What problems do they face in making design decisions?
What information do they need? Or what kinds of assistance or automation would be most useful to support them and improve their designs?
‣
Step 3 - Design
Design a proposed solution using our Generative Design Framework.
Be sure to identify:
The inputs that you’d like to provide to allow users to vary the modeling assumptions.
Which inputs are design variables to be changed and explored?
Which inputs are constants that set up the conditions you’ll be modeling?
The fundamental, underlying logic of the model or system that you’ll implement that relates those inputs to specific, measurable outputs.
The outputs that your tool will provide to its users.
For a configurator, will there be results produced as a Revit model?
For a design assistant, what values or messages will you provide to help guide users in understanding the measures being computed?
For a Generative Design Study, what visualization or summary of the modeling results will help users to easily identify the inputs that yield the most desirable results?
‣
Step 4 - Make your Pitch
Draft up a brief overview of your proposed solution (1 page max) that outlines the essentials of your plan. It should include:
The intended users
The need you’re trying to provide a solution or support for
The inputs
The underlying logic of the model you’ll implement
The outputs
Proposed timeframe/timeline for the various components, what is the minimum viable working tool, which are ‘nice to have’ features
The teaching team will review your pitch posting and provide feedback on your proposal to help guide you through the implementation stage of the process. We’ll share advice about how to best scope the problem and approach it given the limited time available.
Here are some example tools created by previous students:
Precast Bridge Girder Configurator
Ballpark Builder
FABSHEETS: A LASER CUT FILE GENERATOR - Preparing Geometries for Digital Fabrication with Grasshopper
RoofHarvest: generates, evaluates, and ranks rooftop layouts for mixed-use green roofs
TransitCorridify is a generative study tool to analyze the cost, revenue, green space potential, and density score of implementing a transit-oriented development in your city.
Trust this Truss: A Tool for Parametrically Visualizing and Analyzing Pitched Trusses
‣
Part 2 - Implementation Due Fri, June 5
‣
Step 5 - Implementation
Implement your proposed design using:
Dynamo
Grasshopper
Generative Design Tool
And supporting tools as needed
‣
Step 6 - Package and Share Your Tool
Package and share your new tool in a way that allows others to download and easily use it.
For example:
If you’ve developed a tool using Dynamo, set the inputs and outputs to allow users to run your tool using Dynamo Player.
Similarly, if you’ve developed a tool using Grasshopper, use the interface tools to provide slider inputs that highlight the essential values that can be easily changes.
The Generative Design tools offers it’s own user interface, so choose and name the inputs and outputs to make it easy for others to understand and use when you deliver your study. Use the packaging of the Autodesk Samples as examples.
Your complete submission should include:
The Dynamo or Grasshopper scripts and all the supporting nodes needed to run your tool.
Documentation for how to use your tool. You’ll provide this documentation in your Notion posting, and it should include:
A brief overview of what your tools does (to help users decide whether to download it and give it a try). Think of a “ReadMe” file for software.
A catchy name (or acronym) for your tool
A teaser image that shows typical results, i.e. what users should expect to get as an output
A recorded video demo (2 minutes max) in which you demonstrate how a user would interact with and benefit from using your parametric design tool.
If you did not submit the optional Pitch in part 1, this should be included here to show what your original idea was and how it evolved. Please comment on anything that changed, what couldn’t be achieved etc.
Sound intimidating? It doesn’t have to be! The important thing for our class is that you’re exploring how parametric design tools can be applied in a domain that’s meaningful to you.
Think of your initial exploration (for this project) as the seed of a potentially much grander or ambitious effort to apply parametric design tools in truly impactful ways.
For this project, you’ll be creating a first prototype or proof-of-concept -- just enough to test the concept and illustrate how it would work. In other words, a so-called “minimum viable product”.
Don’t be intimidated by this open-ended prompt. Rather, think of this as an opportunity to define your own exploration of a design space that is really interesting to you.
Just let us know, and we’ll be there to support you!
