Model
Generative
Typology
Urban tissue (block & network)
Size
4km x 4km
Fitness criteria
Density, connectivity, open space
“Most studies take the reductions approach of focusing on block morphologies or network structures with no regard to how the two influence, interact or co-evolve over time.”
Co-Evolution
The concept of co-evolution and co-development is not exclusive to living organisms and can be found in the development and evolution of systems of different kinds. Cities, as a collection of differentiated systems that co-develop and co-evolve over time, exhibit similar properties. In modelling interactions between systems of different kinds, the relation between systems, their association and interactions with one another becomes central to the notion of development and evolution of cities over time. The systemic modelling of blocks and networks can be categorised under 3 different scenarios:
Block dominated
Network dominated
Co-evolved
Co-evolved morphologies are more evident in historical patterns of cities as opposed to planned settlements. It is hard to identify a particular pattern type or hierarchy in the development of such tissues, as the evolution of the tissue generally occurs due to many factors during development and at both local and global scales. Unlike the first two scenarios, co-evolved morphologies are not top-down. Hierarchy does not reside within systems (in that a system dominates another, as was the case in the previous two scenarios), but emerges through interactions at the local scale as the model develops.
Computational Modelling of Urban Tissues
Low-density tissue
Gene Count: 14
Network Types: Secondary & tertiary
Network Length: 786
Number of Buildings: 1,681
Solution Time: 36.3 seconds
Area: 2km x 2km
Low-density morphology
The experiment focused on a low-density tissue with a similar morphology to neighbourhoods outside of urban cores. The relational model focused on buildings of small footprints with long-segmented networks. In addition, the building depth is relatively shallow to allow for courtyards. The black and white image has been split in half. On the right is the aerial photograph of west London and on the left is a computationally generated urban tissue embodying the same qualities.
Medium-density tissue
Gene Count: 14
Network Types: Secondary & tertiary
Network Length: 1,804
Number of Buildings: 4,888
Solution Time: 42.3 seconds
Area: 2km x 2km
Medium-density morphology
The medium-density tissue follows a similar morphology as urban residential neighbourhoods of high-density evolved cities such as Tokyo or Istanbul. The building footprints are deeper and the heights taller than the low-density examples looked at. The computational model focused on buildings of medium sized footprints that were density packed coupled short-segmented networks.
Generation: 20
Population Size: 15
Number of Blocks: 690
Number of Buildings: 13,888
Number of Genes: 69,440
Solution Time: 29.4 seconds
Hyper-density morphology
The hyper-density tissue focused on the generation of multiple network and block types. Density measures vary across the tissue as do the morphologies including street and building types.
Hyper-density tissue