Identify populations X kilometers from nearest health facility

[mikefab]

We need to help a government identify the populations most in need of infrastructure and planning for ebola response.

Given the following data sets:

• a list of population points (50k points)
• a list of health facilities (1.4k points)
• shapefile for DRC at admin level 2,
• roads shapefile for DRC

...create a Jupyter notebook that serves all population points x kilometers from nearest health facility.

The final project should be an open-source, standalone GitHub repository, with a README and at least some test coverage. The readme should address (1) what this repo is or does, and (2) how to install and use it.

[radoslawkrolikowski]

Hello. I have been working recently on something similar. You can take a look at my repository here. In a nutshell, I created a map that shows the population (and % of a population) in a district (second sub-national level) that falls within X kilometers isoline of the nearest health facility (or specialist). Isolines can be computed using the following methods:

• Openrouteservice API - up to 2500 isolines a day (100 per minute, max 120 km isolines)
• OSMNX and networkx libraries to create a graph that is used to determine the isolines. The graph can be built from data downloaded using OSMNX or from a shapefile after conversion to OSM format (building the graph can take up to several hours)

If there is still a need for help, then I would like to contribute, either by making any necessary changes to my repository or by creating something similar from scratch that will meet the expectations. I am open to any suggestions.

Sincerely,
Radek