Plans are just adjacency made visible.
Who This Matters To (And Why)
Critical: Architect,Developer,Interior Design. These parties make or lose money directly based on this thesis.
Important: GC,Engineer,City. These parties execute decisions shaped by this thesis.
Context: Banker,Investor,Inspector. These parties need to understand it to avoid friction.
Highest typology impact: Multifamily,Office,Hotel,Mixed Use. Lower impact: Retail,Single Family.
A plan is adjacency made visible. Spatial layout is the physical expression of proximity requirements.
How It Shapes Development
Plans are adjacency made visible because the floor plan is the two-dimensional projection of the building's adjacency graph onto the floor plate. Every wall in the plan is an edge in the graph. Every room is a node. Every door is a traversable connection between nodes. The floor plan is not a drawing of spaces — it is a drawing of relationships. The relationships were specified first, in the program and the adjacency diagram. The plan is the result of embedding those relationships in physical space.
Reading a floor plan is reading a graph. When an experienced architect looks at a plan, they are not seeing walls and rooms — they are seeing adjacency logic. The bedroom cluster in one corner, bathrooms between bedrooms, kitchen open to living, utility accessible from kitchen. This is an adjacency pattern that recurs across millions of residential plans because it satisfies a stable set of use requirements. The specific dimensions and wall thicknesses vary. The adjacency pattern is invariant because the human behaviors it serves are invariant.
Floor plan evaluation is adjacency evaluation. Does the kitchen adjoin the dining room? Does the master bedroom have a private bathroom? Is the mechanical room accessible without passing through occupied space? Is the egress stair accessible from every floor without dead ends? These questions are adjacency questions. A plan that answers all of them correctly is a good plan regardless of aesthetic choices. A plan that answers them incorrectly cannot be saved by aesthetic choices. The adjacency logic is prior to everything else.
Generative design tools produce floor plans by solving adjacency constraints computationally. Given a room list with size requirements and an adjacency matrix with required and preferred relationships, a constraint solver can generate candidate floor plans that satisfy the requirements. The plans it produces look like architecture because architecture is constrained adjacency. The tool is not doing design — it is doing graph embedding. The design was done when the adjacency requirements were specified. The plan is the output of satisfying those requirements within a building footprint.