Buildings account for 40% of global carbon emissions due to the resources required for their construction and operation. While building codes are used to enforce indoor comfort standards and minimum energy efficiency requirements, they disregard the actual spatial configuration and usage of buildings. In our research we show how the spatial usage, and with it efficient building floor plans, are key in creating low carbon buildings. To show this, we present a new computational method and data structure that can represent, analyze, and automatically generate building floor plans, aiming to increase occupant comfort, reduce emissions, and enhance building performance.

The hypergraph – a graph-based representation of an architectural floor plan

We introduce the hypergraph, a generalizable shape generator and descriptor for floor plans. The architectural hypergraph represents key characteristics of the shape divisions of any given floor plan layout, enabling both the mapping and benchmarking of suitable, high-performing floor plans, as well as their automatic generation.  It is specifically designed for building floor plans, encoding relative spatial relationships and geometric properties. Given the same boundary condition, the hypergraph constitutes a bijective mapping that results in the same floor plan and vice versa. This creates a flexible data structure where the same hypergraph can be applied to a variety of boundary polygons to create a floor plan. Different hypergraphs can be applied to the same boundary condition, resulting in different internal subdivisions.

Spatial evaluation and generation of floor plans 

We implemented the hypergraph mapping procedure in an architectural Computer Aided Design (CAD) environment and were able to capture real-world building floor plans. Focusing on a representative set of residential apartment buildings from Singapore, New York, and Zurich, the hypergraph allows us to show and encapsulate architectural differences and investigate spatial configurations that are encoded in local architectural practices, prevalent construction techniques, building codes, and climate.

Through the hypergraph, a 3D model of an apartment can be constructed automatically. This allows for the automated generation of a building energy model (BEM) and a daylight model of a space. Using automatic placement of furniture blocks, we can assess if rooms are large enough to result in livable spaces and compare the overall area to reference floor plans with the same occupancy (Figure 1).

Architectural design of spaces is key for a sustainable building

Our research demonstrates that architectural design significantly impacts building sustainability. Currently, most policy and efforts around creating more sustainable buildings involve incentivizing technical upgrades, additional building insulation or more efficient heating systems. In our research we could show that "space", and through that architectural design, can be much more important when designing low-carbon buildings.

A comparison shows how excess emissions from unused space can be significantly higher than savings from building energy upgrades, especially in the more temperate climate in Zurich (71.6%). The opposite is found in hot and humid Singapore (33.0%) where floor plans are already compact, and envelope performance is crucial due to the climate (Figure 2). This means that in the case of new construction, apartments that are closer to the minimal spatial requirements with less excess space will have significantly lower energy use, even when constructed with less performative envelope standards. Compared to current energy codes that specify performance requirements, our results show great potential in savings through spatial efficiency measures and thoughtful planning and design of buildings – and with it the possibility to include spatial metrics for designing buildings with greater energy sufficiency.

New Opportunities for Architectural Design Workflows and Software

Beyond analysis, hypergraphs can also be used to generate new building layouts. Using a database of hypergraphs collected from real-world building geometries we were able to generate artificial layouts of buildings that were on par – and up to 24% better in daylight performance – than their real-world built reference. The workflow allows for linking of design with building performance simulations (energy, daylight) at the earliest stages of design.

Enabling Low-Carbon Buildings of the Future

We see significant opportunities to apply the hypergraph method to automated benchmarking of building retrofits, including the converting empty office buildings into residential units. Artificially generated floor plans have the potential to improve the quality of residential spaces in terms of environmental performance and access to daylight. Tools like the hypergraph framework can contribute to solving the enormous challenges in building decarbonization and affordability of housing, enabling new creative possibilities in the architectural discipline, through collaborative design tools that can combine machine intelligence with human intuition. Our insights suggest that spatial efficiency and thoughtful planning can lead to substantial energy savings, highlighting the need to incorporate spatial metrics into building design.

A hypergraph analysis framework shows carbon reduction potential of effective space use in housing. Ramon Elias Weber, Caitlin Mueller, Christoph Reinhart. Nature Communications, 2024. https://doi.org/10.1038/s41467-024-52506-z