The DemoLand project

Urban Analytics in the wild

Dani Arribas-Bel

Background

Context

• Newcastle City Council
• Local Spatial Plan
• Policy needs, challenges

Goals

• Quantify competing aspects of land use in a given urban environment through key indicators (baseline)
• Build bespoke scenarios under changes that affect the distribution of such land use (scenarios)
• Present outputs in an interactive visualisation tool

Baseline

Indicators

Air Quality

House Prices

Job Acessibility

Green Space Accessibility

Land Use - Spatial Signatures

Characterisation of space based on form and function designed to understand urban environments

UrbanGrammarAI.xyz

Land Use <–> Indicators

Accessibility scores

• Employment
• Green space

Predictive models

• Air quality
• House prices

What if scenarios

Exploring the effect of changing the baseline land use on the indicators of interest

Baseline

The world as it is

Scenarios

The world as it could be

Co-produced scenarios

• Residential development

  • Low density (1)
  • Mid-density compact (2)

• Inner-city densification (3)

• Brownfield redevelopment (4, 5)

  • Dense neighborhoods (4)
  • Parks (5)

• Combinations (6, 7)

Interactive tool

Demo [URL]

Play time - Activity I

• [10min] Explore the set scenarios on your own
• [10min] Think of a new scenario you would like to explore
  • Why is it relevant?
  • What changes would it contain?
  • What effects would you expect?

Play time - Activity II

New version! [URL]

• [15min] Implement your scenario (use the spatial signature descriptions for help
• [10min] Explore the changes when compared to the baseline
  • Are they in line with your expectations?
  • What is unexpected?
  • Why do you think this is?

“Behind the scenes…”

Urban Data Science and all that jazz…

• Policy relevance
• Data, data, data
• Models, spatial analysis, machine learning
• Co-production

These slides are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.