Sunday, February 22, 2015

Melbourne Car Ownership Map

I am happy to announce the release of our latest visualization developed at the Monash City Science Research Group: Melbourne Car Ownership Map [link].
An important question in urban planning and travel demand forecasting is where households with fewer or more cars live? and why? In this project, we visualize the spatial distribution of car ownership by dwelling in Melbourne metropolitan area using Census (2011) data. Every dot in the map represents a dwelling.
As expected, households who live within CBD and inner city areas have fewer cars, mostly 0 or 1 car per dwelling. As we go further out to the suburbs, car ownership rate increases. 
Click on the map below to re-direct to the interactive visualization.


6 comments:

  1. Looks great, but.....the spatial data needs fixing, I've noticed lots of dots in areas where there are no housing. How can there be a dot (which indicates a dwelling) in an industrial area (such as the Josephs Road precinct around Footscray which is home to a Buddist temple and industrial factories) or in an agricultural area (such as the East Werribee Employment area, which is Melbourne Water land that has no houses currently).

    ReplyDelete
    Replies
    1. Thanks for your comment. You're absolutely right. There are some dwellings that are even mapped on rivers, airport, or parks. The reason is that census data with car ownership info comes at SA1 level. The way we created the map is, at each SA1 level, we randomly generated points (dwelling locations) and if the SA1 zone has a park or river or an industrial area, the points may locate on those areas which is not accurate. Note that we did not have the exact address of each dwelling. All the locations of the dwellings are approximate within the SA1 level. Thank you any way for the feedback.

      Delete
  2. Great piece of work. The data might be more visually appealing and useful if it was presented as a heatmap, showing 'median car ownership per dwelling' at the level of ABS meshblock, with each meshblock being colour-filled, instead of attempting to use dots to represent individual dwellings. http://www.abs.gov.au/ausstats/abs@.nsf/Latestproducts/1386.0Main%20Features3602013?opendocument&tabname=Summary&prodno=1386.0&issue=2013&num=&view=

    ReplyDelete
    Replies
    1. Thanks for your comment. Car ownership info is only available at the SA1 level. ABS only provides people counts at the mesh block level. I believe a dot map is more powerful than a heatmap because it can show the diversity/variety of car ownership at each zone (SA1) while a heat map can only show one single value like the median value as you suggested.

      Delete
    2. It might be possible to get car ownership data from the state government agency (Vicroads) at a more granular level, and dwelling count per meshblock from ABS. This would allow a mean value to be calculated as a representation of car ownership per dwelling per meshblock.

      Delete
  3. Fantastic. Would be great to see it overlayed further with data on incomes as there appears to already be a slight correlation present between low/no car ownership and lower incomes.

    ReplyDelete