Computational Design: A Guide to Implementation

by 

December 30, 2021 | All

If you’re not familiar with computational design yet, it’s an algorithmic problem-solving methodology that uses digital capabilities to develop solutions. A designer can use these digital capabilities to solve design problems (creation, fabrication, interaction, analysis) faster and with more options.

During the December 14 Applied Software webinar, Computational Design: A Guide to Implementation, experts Donnell Grantham, Christopher Riddell, and Anthony Zuefeldt provided an educational overview of computational design.

Computational design involves four steps:

  1. Algorithms – creating step-by-step instructions to solve a problem
  2. Decomposition – breaking a design into smaller parts
  3. Pattern recognition – looking for similarities, trends and patterns
  4. Abstraction – focusing on what’s required to solve the problem and ignoring what’s not necessary

In the webinar, Christopher Riddell stressed that computational design implementation “will not be successful” without the cooperation of leadership. After that, he explained, you can “build strategic alignment” and begin to implement. In implementation, certain steps are necessary: to plan the culture shift that will occur, to democratize usage, and to deploy a development process.

Culture shift:

Successful implementation requires this shift. The status quo of “how you have done things in the past” needs to be challenged in order for your company to grow.  A “communication strategy” should be developed in order to create visibility and spur excitement. Riddell suggested monthly newsletters, lunch and learns, intranets, and hackathons as possibilities.

In addition, Riddell suggested employing the “70/20/10 ratio for usage and adoption” in this scenario. In essence, 70% of the staff should be educated on how to utilize the tools, 20% should have the skillset to conduct basic troubleshooting and editing, and 10% should have the skillset to create future automations and solutions. Add those numbers together, and 100% of the staff should “be encouraged to provide feedback and ideate on future solutions.”

Democratize usage:

Donnell Grantham discussed how to democratize usage, beginning by stating that ease of access is “critically important for long-term success and adoption firmwide.” Adoption can be difficult, so the tools in turn need to be “easy to use, familiar, and relevant.” 

Metric and data tracking are essential to this process, as the ability to review trends will determine if the implementation is running successfully. In addition, building custom design automation tools creates an opportunity to embed usage analytics functions that can generate valuable datapoints and insights into your process. Finally, collecting data can also aid in the learning and development process. Data should be audited on a regular basis, and continuous feedback on the solutions should be encouraged.

Deploy development process:

Anthony Zuefeldt discussed the deployment of the development process, describing it as “an important aspect of implementation because of the impact it has on future tool development.”

The Agile Development Process (ADP), he said, is an “iterative and cohesive method of effective project management and application development that drives the entire lifecycle of the project.” In this process, you first perform feasibility testing and understand context, then you prototype, then you refine and address bugs, and finally you educate and continuously improve your process.

In essence, there’s a lot to learn from computational design and a lot to gain from it if implemented correctly. As long as we continue to ask questions, address concerns, and strive to grow, there’s nothing we can’t accomplish.