At ERA-co, we employ computational and quantitative methods to understand the pulse of places and cities.

While urban data is the raw material for urban analytics, a closer look to the processes involved analysing it can be useful to enhance the generation of information that would otherwise be invisible and provide both a wider vision and a bespoke focus of inquiry.

One of our senior associate data scientists, Nicolas Palominos, looks at the pattern of connections between places in a city, and how they contain valuable information about the way they function. 

This system of connections can be conceptualised as a network in which places are the nodes and the relations between them are the links.

Assuming that there is also data about the size of the nodes, for example, their attractiveness, then the comprehension of places can be greatly amplified through the qualification of the interconnections between them, providing an actionable framework across time, urban space, and urban life.

The interaction between activity nodes can be estimated from their gravitational pull and location. Through the identification of ‘busier and quieter links’, the model provides a quantitative overview of the potential flows (exchanges) between nodes. Movement patterns are analysed through what is technically called a spatial interaction model. This can be considered the first approach to describing the spatial and morphological structure of places and their interconnections. The observation of nodes as the locations where processes of interaction begin and end is useful not only to unveil the pulse of places but also how they function at different scales.

Navigating deeper beyond this somewhat abstract representation of places, after key links are identified, practical applications can be developed. The flows can be mapped into the circulation network in conformity with its spatial characteristics. As a result, the relationship between nodes is captured at high detail providing insights into the way the system and its parts function. Critical pathways along a more detailed circulation network can describe, at a granular level, patterns of activity and behaviour.

The model can be further scrutinised. What are the effects of reducing the size of nodes? Are there significant variations if the choice of pathways change according to the quality of the circulation space? What are the places (nodes) capacities for growth? How can innovation districts, urban campuses, commercial streets, and cities be analysed through spatial interaction modelling? As shown, the placemaking strategies that can be sustained through modelling spatial interactions lend themselves with a rich array of evidence to support a better fit between the dynamics of social behaviour, the built form, and the spatial structure.