Module 5: Spatial Accessibility Modeling

Spatial Accessibility is all about analyzing patterns of how where you live makes a difference in terms of your access to services such as hospitals, fire departments, or food stores. There are various ways to measure spatial access: Euclidean, Manhattan, Network, and Least-Cost Path distances. While all of these metrics are useful, some are more applicable to specific spatial access modeling scenarios than others. For our lab assignment, we practiced measuring spatial accessibility for various populations, first using straight-line distance (Euclidean), and then using network time travel with ArcGIS's Network Analyst Extension. The six most common transportation network solvers in ArcGIS's Network Analyst are: 1. Routes, 2. Vehicle Routing Problem, 3. Closest Facility, 4. Service Area, 5. Origin-Destination (OD) Cost Matrix, and 6. Location-Allocation. Our lab assignment was broken down into three parts, see below for specifics.

Part A: Introduction to Network Analyst

In Part A, we completed four exercises thru Esri's ArcGIS Desktop Help site as follows:

Exercise 1 - We created a network dataset in a geodatabase using San Francisco's street and turn features. We also included historical traffic data to solve time-dependent routes.

Exercise 2 - We found the quickest route (shortest path) to visit a set of stops in a predetermined order, as well as experimented with placing possible barriers.

Exercise 3 - Found four fire stations that provide the quickest route response to a fire at a given address. I also generated routes and driving directions for the firefighters to follow.

Exercise 4 - Calculated service areas (polygons) that represent the distance that can be reached from a facility within a specified amount of time in Paris. Additionally, I created an origin-destination (OD) cost matrix for delivery of goods from a warehouse to all stores within a 10-minute drive time.

Part B: Basic Spatial Accessibility Analysis

In Part B, we did a spatial accessibility analysis using a proximity-based (straight-line) measure. We were provided the following Georgia state datasets: psychiatric hospitals, counties, and census tracts; and then created Cumulative Distribution Functions (CDFs) using scatter plots to determine age-grouped populations with hospital accessibility within a certain distance.

Part C: Spatial Accessibility Using Network Analysis

In Part C, our main goal was to spatially measure drive time accessibility impacted from the closure of a campus at Austin Community College in Travis County, Texas. We created Service Areas analyses of 5, 10, and 15 minutes with all campuses open, as well as omitting the closing campus (subject site). Furthermore, we investigated how this campus closure affected the residents of Travis County aged 18-29. This required a Closest Facility analysis. We were able to quantify how many residents were affected before and after the closure, how much farther they would need to travel to the next closest campus, and the average distance travel increase (9 minutes). And finally, we created a Cumulative Distribution Function (CDF) to compare both scenarios. Below are my final products.