Evidence-Based Optimization in Humanitarian Logistics

Humanitarian crises like the Syrian war, Ebola, the earthquake in Haiti, the Indian Ocean tsunami, and the ongoing HIV epidemic prompt substantial demands for humanitarian aid. Logistics plays a key role in aid delivery and represents a major cost category for humanitarian organizations.

Optimizing logistics has long been at the core of operations research: the discipline that explores the use of advanced analytical methods to improve decision making. The commercial sector has substantially benefited from such methods. This thesis discusses whether and how such methods can also guide policy and decision making in the humanitarian sector. This is done through in-depth analyses of three case studies. The first investigates suitability of advanced planning and routing tools. Next, we investigate decision support methods for designing networks of roadside HIV clinics. The third case study concerns the deployment of mobile teams that screen for infectious disease outbreaks.

Optimization tools come with assumptions about objectives to be reached and about their link with the decisions to be optimized. In humanitarian logistics, safeguarding adequacy of these assumptions is challenging but crucial. Throughout our case-studies, we explore how “best available evidence” can be used to link decisions to objectives, so as to enable evidence-based optimization in humanitarian logistics.