The main concern of logisticians in practicing their trade is to a) determine where resources are needed and b) how to get them there. The secondary clause of their trade is sometimes far easier to figure out: if you need chlorinated water at this cholera outbreak, it's possible to figure out a creative way to get it there. Or if you need 2000 doses of a malaria vaccine at a certain other place, it's possible to fly/ship/truck/motorcycle it there somehow.
But a large part of the medical component of many health projects involves the primary clause of that logistician's trade - define the needs in a geo-temporal state. A decades long trend has been to decentralize this determination - to allow local agents to determine their own needs. But what happens in a system where individuals in a location are mobile? Then who's the 'local' agent and how do they determine needs?
In the modern equivalent of the biologist's migratory mapping technique, Wesolowski et al have expanded on previous work done by Tatem et al to find the dynamic relationship between human migration and human malaria transmission. In their article, Tatem et al used 'anonimized' cell phone usage to track travel from Zanzibar to mainland Kenya and found that a) most people didn't travel out of Zanzibar and b) that those who did generally travelled to low malaria prevalance areas, such as Dar es Salam. (In a major anti-privacy fear, Tatem et al do acknowledge that: "risk groups visiting higher transmission regions for extended periods of time could be identified" - this fear probably deserves its own post later).
Expanding on Tatem, Wesolowski used 'anonomized' cell phone calls to track individuals ALL over Kenya. After mapping the migrations of every individual phone number, and therefore the corresponding individual, they overlayed this information to find the 'sources' and 'sinks' of malaria transmission - the sources being where someone got infected and the sink being where someone dropped it off:
I don't have the technical chops to determine if the calculations are accurate. But it seems that our mapping skills are advancing far ahead of simple cartography and that these exercises are reasonable flexings of our technologic prowess. If so, then the concepts of decentralization and logistics are, in turn, greatly advanced also. If a key question in decentralization is determining logistics needs in a dynamic needs environment and if a central question of logistics is figuring out the micro-level needs, then mapping exercise like these can go a far way in providing the answers that logisticians need.