1. Business analytics, as the words suggest, is mainly used in commercial environments, but the problems it can solve also exist in humanitarian fields. What brought you to harness business analytics to address the food problem?

"I had always worked on optimizing complex logistics for large companies, until about ten years ago someone asked me whether the same solutions would not also work for WFP, the UN's World Food Programme. That question fueled my personal desire to make the world a better place with our field.

Around the world, 811 million people go to bed hungry, with WFP providing food to about 125 million of them every day, in more than 80 countries. It’s a complex operation, because you have to figure out how to feed very different groups of people while minimizing overall costs in order to help as many people as possible.”

2. You tackled the problem with a small team. How successful has this approach with business analytics and operations research been?

"There were two problems: the diet problem and the supply chain problem. We developed the Optimus model to address both issues at once. One of the innovative features of the model is that it is not based on food products, but on nutritional requirements, which makes the model a lot more flexible because it allows users to consider alternative foods that are closer by, for instance. Ultimately, it means WFP can provide the same nutritional values at a lower cost. According to WFP executives, the use of analytics has saved $150 million so far; enough to feed two million people for a year. We can honestly say that our field is making a substantial contribution to reducing hunger."

3. Are there other possible applications in the field of humanitarian aid?

"We have also been thinking about how analytics could help other nongovernmental organizations, prompting the launch of the Zero Hunger Lab and, more recently, Analytics for a Better World. Through these initiatives, we aim to use applied mathematics to help achieve some of the UN's Sustainable Development Goals.

“We can tap into our creativity and help them with algorithms, but the best models are made in close collaboration with the organization you’re working for. The more domain knowledge you acquire, the better the models become, and the more people benefit. We’re currently exploring ways to use image recognition to detect malnourished children. Could we detect malnutrition simply from a photo or video of a child in Africa or South-America? If we could, the impact would be tremendous, because it would save families from having to visit a health center two villages away, as well as saving doctors and nurses a lot of time.

I’ve also partnered up with a colleague from the Faculty of Law to work on an SDG initiative in the field of refugee flows. 66% of refugees live anonymously in cities, and when we reached out to these so-called urban refugees in Greece, we managed to find 300 people in two months to ask them more than 100 questions. In the hands of a data scientist, the resulting information is very valuable, as it taught us how people in these cities survive and whether they experience extortion or labor or sexual exploitation.”