The location and location-routing problem for the refugee camp network design

The refugee crisis is one of the major challenges of modern society. The influxes of refugees are usually sudden and the refugees are in sheer need of services such as health care, education and safety. Planning public services under an imminent humanitarian crisis requires simultaneous strategic and operational decisions. Inspired by a real-world problem that Red Crescent is facing in Southeast Turkey, we study the problem of locating refugee camps and planning transportation of public service providers from their institutions to the located camps. Our modeling approach brings a new facet to the location and routing problem by considering the location of beneficiaries as variables. We develop a branch-price-and-cut algorithm for the problem. To solve the pricing problem, we introduce a cycle-eliminating algorithm using nested recursion to generate elementary hop constrained shortest paths. The best version of our algorithm efficiently solves 244-node real-world instances optimally.