Traffic Analysis Zone Creation from Census Blocks

2 posts / 0 new
Last post
David A Nelson
Traffic Analysis Zone Creation from Census Blocks

Greetings Colleagues,

NMDOT is performing initial work for updating the State Travel Demand Model. Most data has been acquired and current work is headed toward creation of a new TAZ structure. For the most part, census blocks with relevant attributes will be aggregated into TAZ. We are currently loading census attributes into blocks and will add data from residential and commercial points in the future.

Two challenges exist for moving forward. 1) Loading summaries of point data into individual blocks. 2) Aggregating the blocks into TAZ sensibly.

The first challenge is likely achievable through use of our Arc Desktop GIS, but it would be nice to have a dedicated tool/extension which makes this very easy. ArcGIS Pro has a tool which look promising: Aggregate Points.

The second challenge involves automating aggregation of blocks into TAZ by adjacency, homogeneity (population/employment), and a population + employment formula (min,max,target). The result would have to be checked and compared to physical features such as road network, railroads, water bodies etc. before accepting the aggregations. Something which quickly moves a block from one proposed TAZ to another would be great.

Any thoughts you have are most appreciated.



Hi David,

A few years ago our team went through the process of aggregating census blocks to micro-analysis zones (MAZs) which were then nested into TAZs. This is in a nutshell what I would do if I were in your shoes.

I would not work with  census blocks at all since in a lot of cases their boundaries are not meaningful (they were not reviewed by humans). I would instead start with census block groups and "drop" roadways (preferably future roadways ), hydrology networks (natural and man-made) on top of them to cut them (subdivide) them. This would give me draft TAZs - version 1.Then I would calculate the population, employment, college enrollment etc. from loading point data in the draft TAZs.  I would select the polygons that exceed my thresholds and review them. Wherever necessary I would subdivide - this would give me draft TAZs version 2. Then I will recalculate and subdivide again. It would take a few iterations but it would be faster than grouping blocks.

 Subdividing polygons is a mix of automatic procedures, review and manual splits. Loading attribute data from points to polygons  is a common GIS funtion.

It is inevitable to fall bellow the minimum values for population, employment etc. in  scarcely populated areas and to exceed the maximum employment threshold in a major job center.

Hope that helps,