# The Nature of Human Settlement: Building an understanding of high performance city design

arXiv: https://arxiv.org/abs/1910.03219 · HTML: https://mlsanity.com/paper/1910.03219/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Oct 2019.

## Built on by (most cited first, 1 of 1)

- [Self-Supervision, Remote Sensing and Abstraction: Representation Learning Across 3 Million Locations](https://mlsanity.com/paper/2203.04445/) (2022)

## Abstract (arXiv)

In an impending urban age where the majority of the world's population will live in cities, it is critical that we improve our understanding of the strengths and limitations of existing city designs to ensure they are safe, clean, can deliver health co-benefits and importantly, are sustainable into the future. To enable this, a systematic and efficient means of performing inter- and intra-city comparisons based on urban form is required. Until now, methods for comparing cities have been limited by scalability, often reliant upon non-standardised local input data that can be costly and difficult to obtain. To address this, we have developed a unique approach to determine the mix, distribution, and composition of neighbourhood types in cities based on dimensions of block size and regularity, sorted by a self-organising map. We illustrate the utility of the method to provide an understanding of the underlying city morphology by overlaying spatially standardised city metrics such as air pollution and transport activity across a set of 1667 global cities with populations exceeding 300,000. The unique approach reports associations between specific mixes of neighbourhood typologies and quantities of moving vehicles (r=0.97), impervious surfaces (r=0.86), and air pollution levels (aerosol optical depth r=0.58 and NO2 r=0.57). What this illustrates, is that this unique approach can identify the characteristics and neighbourhood mixes of well-performing urban areas while also producing unique `city fingerprints' that can be used to provide new metrics, insights, and drive improvements in city design for the future.