Optimization for reducing hunger and improving healthcare access

Information Systems and Operations Management 

Intervenant: Dick den Hertog (U of Amsterdam)

Salle Bernard Ramanantsoa

Abstract

At Analytics for a Better World (https://analyticsbetterworld.org/) we use and develop analytics techniques to contribute to the Sustainable Development Goals. In this talk I will describe two of such applications.

The first application is the optimization of the food supply chain for the World Food Programme (WFP). For each new operation, the WFP has to decide “which ingredients should be in the daily meal, and how much, such that all kinds of nutrient requirements are satisfied”, “where to purchase those ingredients and how to transport them to the beneficiaries, such that total costs is minimized”. I will show how optimization is used to answer these questions. Due to the use of these models and techniques, WFP can feed millions more people.

The second application is about maximizing geospatial accessibility to healthcare, which is extremely important for good health and well-being.  In many low and lower income countries the access to healthcare is low. We developed methods to answer questions as: “What is the percentage of the population that has access to a healthcare center within 5km walking distance? How many new centers and where should we add such that this coverage percentage is increased to 95% ?”.  Such questions can be modeled as facility location problems, which are well studied in the literature. However, since the resulting facility location problems for geospatial access are huge, we developed new methods. During my talk, I will discuss these new methods, and also the application to optimize primary health care facility locations in Timor-Leste, stroke center locations in Vietnam, covid-19 test center locations in Nepal, and water well locations in Sudan. These projects are carried out in collaboration with NGOs, as the World Bank, the World Health Organization, Amref, and the Red Cross.