If you read the papers you could be forgiven for thinking that the end of the world could come at any moment. Dangerous politicians, climate change, ISIS, aliens, Zika, microbeads in shampoos - they’re all apparently coming for you. But although these may all feel like clear and present threats, the most likely cause of the apocalypse is still zombies.
Many of you will have started packing your zombie survival kits already. Items you’ve chosen so far will likely include a spear, a gun, a bag full of slim jims, a copy of Queen’s seminal 1976 album ‘Day At The Races’, and a couple of Where’s Waldos. If this is the case, you have chosen wisely. However, an even more vital tool in your belt, necessary for your very survival, is analytics. We’ve looked at how it can help you at every step of the way.
1. Predict the spread of the virus
In the early days, it is likely that many vital resources - the internet, electricity - will still be available. At this stage, the onus is not so much on survival as it is prevention. As with all contagions, one of the most pressing challenges for its containment is understanding where it will spread. Infectious disease physician at Toronto-based St. Michael’s Hospital, Kamran Khan notes that one thing is true of the spread of infectious diseases: ‘If you start to analyze the situation when an outbreak occurs, you’re already too late.’
The most important thing to establish first is where the virus is likely to spread next, and analytics is vital in that. Analysis of data streams like flight itineraries and population density can be blended with clinical information to better understand the points of origin and where it’s likely to strike next. Our experience with the Ebola virus - often called the zombie virus because of its affect on people - and Zika has already taught us a great deal about predicting where viruses are spreading and what to look for, which should stand us in good stead.
2. Find a cure
While evacuation and containment is the main priority, scientists should simultaneously be looking for a cure. According to a recent paper, ‘Use of machine learning approaches for novel drug discovery’, machine learning can now be applied through several stages of drug discovery methodologies. These include ‘the prediction of target structure, prediction of biological activity of new ligands through model construction, discovery or optimization of hits, and construction of models that predict the pharmacokinetic and toxicological (ADMET) profile of compounds.’ For example, an AI-led experimentation system can identify which experiments to conduct during the testing. It finds patterns in the data to accurately predict results of experiments without actually carrying them out, and cuts the amount of tests required by as much as 70%.
3. Where to run
While analytics is proven to be highly effective when it comes to predicting the spread of disease and finding cures, it is not flawless. An algorithm from Nashville-based health analytics firm WPC Healthcare was able to predict the spread of another virus spread via mosquitos, the West Nile virus, with 85% accuracy, but there have also been costly failures. Google’s Flu tracker, for example, was a spectacular failure that is often held up as a warning of the hubris of data practitioners. As anybody who has seen zombie films will know, cure and containment usually does not work. If it did, I Am Legend would likely have been incredibly boring, well, more than it was.
So, your attempts to stop the virus spreading and find a cure have failed. You’ve survived, along with a band of others, and you’re on the run from the zombie hordes. Hopefully, you’ll have gained some insights as to where to go by looking at the spread of the disease, and you can head towards the least populated area. If not, people from Hopper.com have already used data to get an idea of the best places to go. They created a data set ranking each state's survivability, looking at key metrics including gun ownership (to kill the zombies), number of Walmart stores and hospitals for supplies, number of military bases for protection, population density and number of urban centers. They also looked at the best kind of terrain, with mountainous states ranking highly. They named Texas as the best place to head, noting among its strengths its low population density, high number of gun shops, hospitals, military bases, and Walmarts in comparison to other states. DO NOT go to Rhode Island. According to Hopper.com, Rhode Island is comfortably the worst place to ride out the zombie apocalypse, with high population density, lack of mountains, strict firearms laws, and little else.
4. Better understand your human capital
By this point, the internet will be gone and social media will likely be down. While some might say that without social media, there’ll actually be less zombies, but without it, it will certainly make gauging sentiment around a fightback tricky. It is, however, pretty safe to assume that people’s feelings towards flesh eating zombies will be almost uniformly negative, so finding an army and getting people to pitch in shouldn’t be too irksome.
The key to creating your new society is mobilizing the human capital available for effectively surviving, understanding the people you have at your disposal the functions you need filled, and what qualities would best be applied to ensure the best people are utilized properly.
This does not need to be done digitally, small data is equally impactful. Questionnaires should be handed out to everyone so you can determine metrics such as skill sets, work ethic, and ways people like working so that you can help them realize their full potential by creating an environment where they can do their best. Most likely, somewhere they won’t get eaten. Ordinarily, in the zombie-free paradise we live in today, you would use a cloud-based software system to scan applicants’ online assessments and determine how closely candidates’ behaviors align with profile of the role you’re employing for. In a world without the internet, it’s back to basics, and drawing up spreadsheets with a pen and paper. It’s definitely slower, but you’ve got time.
5. Pinpoint the zombies’ weaknesses
Now you’ve got your new society, you need to defend it. Hopefully, thanks to your understanding of human capital, you have your army. Now you need to understand your enemy. Marketers have been using behavioral analytics to understand their customers, but applying such techniques to zombies is unlikely to yield results. The insights garnered will be on a smaller scale without large scale computing to help you, but there is still information that you can gain.
You have the insights already garnered by the various militaries around the world to fall back on, and you can have everyone note down their experiences of the zombies for analysis to better understand how they operate. You can even set up more primitive data points to assess their movements - traps for example, to measure how many times zombies go down a certain path. A foghorn attached to a piece of string for example, that lets you know whenever a zombie triggers it, will give you an idea of where they’re heading. You could even leave various foods out and play different music and note any increase in the number of zombies walking down that path to see what triggers them. Understanding your enemy is as vital to surviving the zombie apocalypse as understanding your customer is in marketing. The tools may, by necessity, be less advanced, but the same principles apply, and learning as much about analytics as possible before the biting starts may be your only chance to survive.