I'm trying to build a decision support model for a brand to help in deciding where to open a new store of the brand in an urban area.
The model will be focused on location observations (lat-lng, timestamp, duration) of smartphone users around the city (without income and finance considerations). I've extracted from these observations visiting occurrences at stores of the brand and also other brands around the city.
I'm trying to plan my next steps but unfortunately right now I'm a bit stuck. I was thinking to try a network approach to find a missing node, but still defining the connections is still tricky.
Another example: to do cluster analysis and profiling the segments of users that are visiting in the successful branches, and then project the results on a different area in order to find a possible missing branch, where the specific segment of users exists.
I would appreciate if someone encountered a related paper that might shed some light on how can I progress, or maybe have an idea of how to move on?