Which GIS best for this problem?

Question

This problem is similar to retail presence optimization. I'd like to know which GIS's could help us do the following:

Given a set of point locations,

1. create a circle of radius n meters around each.
2. each circle has a unique score.
3. replace the circle's unique score with the sum of its score and the scores of all circles with which it intersects.
4. create a polygon(s)--the union of all circles with sum >= a threshold.
5. Layer in a new set of points and take the set of all these new points that are contained within the polygon(s).
6. Export as a table those points and their metadata.

I once upon a time was reasonably fluent with Manifold, and I think this could be done in Manifold, I think. We're interested in a GIS, for now, costing no more than Manifold. Shortest learning curve super appreciated too.

Thanks in advance for any guidance you can provide, Steve

Answer 1

These activities should all be easily done in QGIS, point by point as per your question:

1. Symbols can be given radii based on a field. In the Symbology properties tab, select 'advanced', 'size scale field' and then the field with the radius (the CRS of the project will need to be in metres).
2. The score can be stored in an attribute field;
3. "Join attributes by location" can be used to summarise intersecting layers, probably with a subsequent field calculator operation to total the scores;
4. Performing a query (Layer|Query...) on the circles layer will let you select only the characteristics you want, e.g. sum > x, and then Vector|Geoprocessing|Union to create a new layer with those circles.
5. A point-in-polygon operation will give you the interesting points within the polygon layer.
6. The attribute table can be saved directly with save-as CSV.

And, as we love to point out on this site, QGIS is FREE.

Answer 2

For the described issue I don't see the point in actually using GIS (you didn't mention that you have to visualise data). This issue (especially if it is routine one) can be solved by several quite simple SQL-statements. Consider using PostGIS or Spatialite (depends on your data amount). I believe another alternative is to use OGR scripting but I would definitely stick with database management system.