ProjZim

Discussion

 The use of Volunteer Geographic Information along with our own manually digitized data, while being easily accessible, did come with several drawbacks. For our digitized roads we had set criteria on how we wanted each type of road to be defined, however we ran into many inconsistencies with the previous road network data that have been already digitized before us. We were also hoping to create a dataset for the rural settlements found within our study area as we worked on digitizing the road networks. However, we quickly realized that without any form of ground truthing, it was incredibly difficult to identify which buildings in the satellite imagery were in fact houses. This approach of utilizing VGI must be used with caution in order to avoid faults in the anlysis due to data quality.

Limitations

 There were a number of limitations to this project that we ran into. Firstly, due to time constraints and the overall lack of data for Zimbabwe we were unable to conduct our study over a larger study area. This would have created a much more comprehensive analysis of the state of healthcare for Zimbabwe but having to manually digitize the majority of the roads within the country, this was not feasible. Originally we were hoping to work alongside a group of students within Zimbabwe that would aid us in digitizing the roads as well as ground truthing some of our data but due to scheduling we were unable to carry this out.

 We were also hoping to include travel time in addition to distance as another method of analysis. This would take into account road conditions as constraints into our analysis. It would take longer to travel along dirt roads than paved ones. However our OSM street data that contained the information in order to do this type of calculation would not properly load into the "Network Analyst" tool within ArcGIS. To solve this we ended up having to convert our road network polyline into a plain polyline shapefile and in this process all of our attribute data was lost.