Analytics for a Better World

For the second year in a row, our team-member Sophie Caro attended the Analytics for a Better World conference on May 14th. Nowadays, analytics play an important role in increasing a company’s success. ‘Analytics for a Better World’ envisions that analytics can similarly make an impact for societal and humanitarian challenges. They dream of using analytics to further support the Sustainable Development Goals (SDGs).

Their event brings together speakers and panellists from different groups: INGO, NGO nonprofits, UN organisations, social enterprises, researchers, and companies. These organisations presented several application cases for analytics in nonprofits and how analytics can help make the world a better place. This blog will showcase four use cases that Sophie was able to learn about during the event.

1 – MSF’s Malaria Anticipation Project

Medecins Sans Frontieres (MSF) uses predictive models to provide an “Early Warning Early Action” (EWEA) system for malaria transmission. Malaria is endemic throughout South Sudan. With high transmission rates throughout the year and seasonal peaks around September to January, it is one of the leading causes of child mortality in the country. They use environmental indicators to identify variations in malaria transmission. The EWEA system allows acting before a threat to reduce impact on the population. For instance by increasing clinical capacity or implementing prevention activities. By doing so, MSF uses analytics to reduce malaria morbidity in South Sudan! Check out this link to read more

2 – Estimating child poverty in Sub-Saharan Africa

Save the Children is using Machine Learning to estimate where in Sub-Saharan Africa children live in multidimensional poverty. Multidimensional child poverty means that children are severely deprived of at least one dimension of sanitation, water, health, nutrition, education or housing. Globally, it is estimated that over 1 billion children live in multidimensional poverty. Making them twice as likely to pass away before they grow to become adults. Mapping child poverty allows governments and other organisations to effectively design policies and fight child poverty.

However, reliable data is scarce and for some countries, the only available data is at country level. In this project, publicly available data sources such as satellite images, demographic and economic data are used to estimate child poverty. Specifically, they provide micro-estimates of prevalence, depth and poverty dimensions at a resolution of 5.2 squared kilometres. Using this data, governments and organisations can more effectively fight child poverty. Check out this link if you are interested in learning more.

3 – Optimising investments for Dutch food banks

Researchers from Tilburg University have used data analytics to build an optimisation model that prioritises investments that will have the greatest societal impact, defined by the number of people that can receive food assistance. The Dutch food bank struggles with shortages in transport, storage and food donations. Hence, decisions have to be made regarding which shortages require investments. Using real-life data from the food bank supply chain, they were able to establish how to effectively invest in transport, storage and food donations. These effective investments help serve 32% more beneficiaries and increase the food bank’s capacity. The association of Dutch food banks has already started applying these findings. Want to read more? Check out their article.

4 – Optimising people’s diet to increase access to food

The World Food Program (WFP) tries to optimise people’s diet by finding the most affordable combination of available food that meets their nutritional needs. This optimization is a Linear Programming problem where several dimensions are taken into account: price, availability, nutrition and diversity. Through efficient, optimised diets, more people get access to food. According to the WFP, they have made 150 million USD with data science, allowing them to feed two million more people for a whole year. These solutions originate from a more corporate environment, where Tilburg University professor Hein Fleuren previously built similar solutions. To read more about his story, transitioning from a corporate environment to actively contributing to a better world, check out their blog.

All in all, these four projects clearly showcase how analytics can contribute to a better world. Predictive models empower MSF to tackle malaria outbreaks through an “Early Warning Early Action” system. Machine Learning, combined with public data, allows Save the Children to pinpoint areas of extreme child poverty. Dutch food banks leverage optimization models to maximise their impact, reaching 32% more people with food assistance. The World Food Program utilises data science to optimise people’s diets, also enabling them to feed more people, millions more even. At The Data Story, we value societal impact and making the world a better place. We hope this blog has inspired you to also use your skills in bettering the world.

Follow The Data Story on LinkedIn to stay up to date on our blogs and more!