In collaboration with NASA and INVAP our team developed a solution that uses machine learning and artificial intelligence to predict which groups of people may be more vulnerable to environmental hazards. The solution consists of 2 layers, the first one focuses on predicting the areas that are prone to suffer from natural disasters creating a risk index depending on natural factors. Then, the second layer uses the calculated index in conjunction of probational and geographical data to estimate which parts of the population are more vulnerable to these events.

The first challenge was the lack of time, because of this we have chosen to focus in one country and one particular hazard that affects it in a significant way. For the first layer, we used many of the NASA weather satellites being able to use multiple data sets as for example soil moisture mean and ground temperature.Then for the second layer map we used the we used the local governments census data to pin point the population and the most significant information calculating the vulnerabilities according to our parameters like economic level, number of hospitals per habitat, etc.

with the different skills our group managed to create an escalable solution, that may help to save millons of lifes.