@aurizon. That's a very intresting idea. My thought on this was to use data from opioid overdoses, opioid deaths, opioid arrests to identify goegraphic cluster

@aurizon. That's a very intresting idea. My thought on this was to use data from opioid overdoses, opioid deaths, opioid arrests to identify goegraphic clusters. Then isolate USPS data for parcel delivery from China to an address that lies within the circumference of the geographic clusters. Filter on optimizing factors and features - higher volumes and occupants of those addresses that are known to be associated to illegal drugs (prior arrests etc.). Then investigate those packages.