Modelling air traffic to slow epidemics

By » Fri, May 18 2012
Lost Links and Mid Air Collisions The Problems With Domestic Drones 450x277 Modelling air traffic to slow epidemics

"This is clearly an important species we're dealing with, and I don't think you or I or anybody has the right to arbitrarily exterminate them." (img by DoD)

A couple of academics at the Newcastle University, Jose Marcelino and Marcus Kaiser, tackled the very question that keeps us all up at nights, “How effective would shutting down air travel actually be in an epidemic?” The answer, according to their models, is lots.  Lots effective.

The problem is that it would choke society as much as it reins in the spread of disease. What’s way more interesting is that they were able to achieve the same level of spread reduction by grounding just one in five flights, isolating certain hops based on their likeliness to spread contagion globally.

They found that shutting entire airports only had a significant effect on spreading if it reduced travel by 95 per cent. By contrast,  they could achieve the same effect by removing just 18 per cent of flights between cities ranked by a network measure called edge betweenness.

At best shutting entire airports could only cut infections by 18 per cent whereas removing specific flights reduced infections by up to 37 per cent.

“Selecting highly ranked single connections between cities for cancellation was more effective, resulting in fewer individuals infected with influenza, compared to shutting down whole airports,” say Marcelino and Kaiser. This approach has the added benefit that it disrupts far fewer individuals.

Eventually, everyone will be exposed to the pathogen, that’s not in question. Even you, Japan, with your hand sanitizers and around-town face masks. And God help you ’cause you know your immune systems are shitier than shit.

This dovetails in nicely with a recent set of models by Fuck Yeah Molecular Biology tackling different types of zombie apocalypse scenario, based on several zompoc contagion precepts. (By the way the search term “zompoc precepts” yields delightfully few results.) Long and short: you better hope there’s a cure. 

F9JTm Modelling air traffic to slow epidemics

This is our basic model outbreak scenario, and it doesn’t look good. Zombies quickly take over, infecting everyone. Ouch. (img by Fuck Yeah Molecular Biology)

If we assume Wikipedia’s initial postulates about zombie apocalypse to be correct, as outlined here, and we assume that zombies are created by a microbial agent that will eventually remove them after a given amount of time (otherwise, after all the humans die we’d have to get into the population dynamics of zombies, which would be pretty useless seeing as we’d all be dead anyway. So let’s give us a fighting chance, shall we?) We’re also going to assume that humans who encounter zombies will enter blind panic, leaving them totally susceptible – or naive – to zombie attack, and blood-to-blood pathogen transmission, just because it’s awesome.

Based on these initial conditions, we can construct a mathematical model for a theoretical zombie invasion of Earth. This will be loosely based on the epidemiological SIR model, but will have distinct differences in that R will no longer stand for recovered, but for removed, and there will be a latent period of infection for zombie characteristics to take hold of an infected human. For simplicity’s sake, we’ll also assume the R0 of the zombie pathogen is 1.

And since I don’t like the precepts used to establish the type of zombie scenario where these models yeild survivors, I have to come to terms with some math I found on the Internet being wrong, or still worse, that I would not survive a zombie apocalypse, no matter how well I can swing an ax.

That being said, society benefits from molecular biology enthusiasm, so head on over to Fuck Yeah Molecular Biology and read the whole thing.

Bonus points if you have any idea what the non-graph business is about.

Related: Popular:

Features, Know, Live