Our platform is based off of real data that is available to the public. We, as a team thought it would be more interesting to work with real life data, rather than "Dummy Data". There were numerous steps that were followed to produce this information and the detailed process was presented during the second demonstration.
Just in case you've forgotten the process, here's a quick summary:
We started off with the catchment area rings for each healthsite derived in Maptitude and the population aggregated into wards. These two layers were overlaid and then the percentage area of the ward that overlapped with the catchment area was calculated. The proportion of population of the ward equivalent to the overlapping area was then assigned to the catchment area. Finally, these values were summed to generate the total population in each catchment area.
As said above, this process has a few limitations and assumptions - lets go through them:
Outdated Data
The population data is based off of the census 2011 data - that's almost 10 years ago and a lot has changed since then. The roads network data used to generate the catchment areas in Maptitude is from 2013, more recent data with enforced topology was not available.
Population not evenly distributed
It is assumed that across a ward, the population density is constant, however this is not the case in reality as numerous factors such as uneven terrain prevent this from happening.
People will go to the healthsite nearest to their house
This project is based off the assumption that people will go to their nearest clinic and therefore if they reside within a catchment area they will go to that specific clinic. This, however is not the case in reality, as people may visit a clinic that has better facilities or is on their commute route to work or school.
Don't account for overlapping catchment areas
When the catchment areas were derived in Maptitude, there were overlaps. Nothing, however was don't to avoid this and therefore some areas have a population that is counted twice. This is a limitation, however we are also aware that some people travel further than the specified 1500m and therefore except this limitation to cancel out the duplicate areas.
People will walk at least 1500m to get to a healthsite
This is an assumption, however this is not always the case and the threshold distance would vary according to person.
We have all the healthsite info
Our final assumption is that we have an exhaustive datasets of the healthsites in the Mamelodi area. Should there be other clinics or mobile field hospitals, all the information derived from this project would be flawed.