Moving Towards a Physical Typology of the City

Happy New Year to everyone!

Note:
The thoughts laid out below are very preliminary and unpolished, my hope is to engage in an evolutionary conversation to further refine these thoughts.

Introduction:
Starting off 2010, I wanted to focus my attention on a question that often comes up as part of my daily work. Regularly, I am asked to create segmentation of markets to help prioritize investments by clients. The standard approach has been to identify either economic or demographic segments (i.e. typologies) and use that to draw inferences on the type of investment that might be necessary.

Missing an important piece of the spectrum:
This approach generally yields results that work to understand the needs of certain customer types and the “treatments” that might work for those customers. Unfortunately it misses a very important piece of the spectrum, namely to understand how the shape of the city can catalyze or impede the success to serve/acquire those customer segments.

For example, take two fairly similar cities such as Boston and San Francisco. While San Francisco is larger than Boston, these two cousins cities generally have the same socio-economic profiles (highly educated, low unemployment rate, similar primary industries, similar industry make up – all from census.gov) [Future entry: finding similar cities], but these cities are actually quite different in their physical shapes and structures, as such, a bank or retailer may need to think about their network configuration in those two cities very differently.

Why?
What do we know about San Francisco and Boston? San Francisco has many neighborhoods and pockets of population density. It also has a few major arteries that connect the various nodes of the cities. Finally, San Francisco is a “presque-isle” – a piece of land that juts out into the ocean.

Boston on the other hand, is very well connected to neighboring cities, and has different pockets of population density. Boston’s neighborhoods tend to either be larger (the North End), or string along major corridors (Allston).

This becomes important in thinking about one’s network density, especially in those industries where physical convenience is a factor. In those industries, the general axiom is that, up to a certain point, the closer same brand stores are to one another, the better it is for that stores bottom line. Conversely, the further stores are apart, the harder it is for those stores to maintain a strong brand awareness. But why would one of those companies put a store in an area where there are no consumers or businesses to serve.

Additionally, if the population in a given neighborhood is highly concentrated, then what decisions should be made about the shape of the stores in that location. (BTW, this is generally well understood).

My point is simply that the physical shape of the city does a lot to affect the performance of a network of stores.

How would I go about defining it.
So how would one go about measuring the shape of the city. For this, I would propose an approach that borrows from Kevin Lynch, Christopher Alexander and Richard T. T. Forman.

If you consider that the city is an ecosystem where people have a habitat that they thrive in and need to access food, you can start to think about the distribution network of a given company as a set of nodes, corridors and islands.

nodes are small locations where there might be one or two stores and very few people. This could be an up and coming neighborhood.

islands are different than nodes in such that they are not only larger, but also have a certain degree of self organization and structure within them. For example, San Francisco’s Chinatown is a form of island, as there is a cultural elements that self organizes that neighborhood.

Corridors are the arteries that connect nodes and islands. Generally these are the major commercial thoroughfares of a the city, but can also be the residential boulevards such as Comm. Ave in Boston.

A physical typology of a city would characterize each city by the number, size and shape of nodes, corridors and islands.

More to come on later posts
Thread: City Typology