[Generative Design] Why "Rules" Matter More Than Algorithms: A Case Study on Water Treatment Facilities

Integrating Human Design Intent into Automated Layouts (feat. CAD to Revit Workflow)

https://www.linkedin.com/pulse/layout-review-wonho-cho-ubjgc

Introduction: The Missing Piece in Automation The central water treatment process varies depending on the situation, but the flow is critical according to the exact plan. When implementing integrated flood control plans, we often turn to Generative Design (GD) to find the optimal layout. However, a common pitfall exists: raw algorithms often produce results that are mathematically "correct" but practically useless.

In this post, we review a case study conducted with Kunhwa Engineering, exploring how to bridge the gap between algorithmic generation and human engineering judgment using Dynamo and Revit.

1. The Workflow: From CAD Lines to Dynamo Logic The process begins with legacy data. We separated the CAD area of the alternative into site boundaries, spaces, and facility areas.

CAD > Revit > Dynamo: instead of drawing from scratch, Dynamo recognizes simplified information inserted into Revit.

Distance Calculation: Simple distances are calculated according to the process order. Later, pipe lengths and connecting distances can be extracted and filtered separately.

2. The Challenge: Why Pure Randomization Fails Initially, we defined the adjacency of main spaces as rules and compared GD results against existing alternatives. However, without specific constraints—such as facilities required near the entrance or traffic flow—it was difficult to extract alternatives that a human engineer would judge as suitable. The algorithm didn't "know" where the entrance was.

3. The Solution: Defining "Human Intent" as Constraints To fix this, we created explicit constraints. For example, we set a rule to position specific spaces (Red #1 and #2) within a recognized area close to the entrance.

Constraint Logic: By adding these constraints as rules, we confirmed results similar to the alternatives a professional designer would consider.

Advanced Features: Beyond 2D layout, we added circular circulation paths and even extracted cross-sections by assigning height values and volumes to the arranged spaces.

Conclusion: Generative Design is a Language, Not a Magic Wand The most important implication of this study is that Generative Design yields meaningful outcomes only when human design intentions and experiences are clearly defined as 'rules' and digitally applied.

Human Process as Rules: GD results must incorporate the planning process.

Clarity: Constraints should be set as clear options.

Autonomy vs. Accuracy: Flexible adjustment of location variables is key to achieving both.

This approach reduces simple repetitive tasks and helps designers focus on more creative and strategic decision-making.