A robust optimization approach to volunteer management in humanitarian crises

In the aftermath of a disaster, humanitarian organizations quickly assemble a workforce that can immediately serve a community׳s needs. However, these needs change over time, and the volunteer base (and their skill sets) also changes over time. In this paper, we develop a flexible optimization framework to dynamically allocate volunteers in order to minimize the cumulative unmet demand and maximize volunteers׳ preference. We use a robust optimization approach to handle the uncertainty in task demands because in the scenarios of interest it is unlikely that demand distributions are available for decision makers. We consider maximizing volunteers׳ preference by introducing a constraint into the model which enables decision makers to derive Pareto optimality and allocation decisions for any degree of conservativeness. Our numerical results show that volunteer managers should consider matching volunteers to their task assignment preferences up to a critical percentage, above which needs fulfillment decrease quickly due to overly strict adherence to volunteer task assignment preferences. Moreover, one can estimate the complete price of volunteers׳ preference by a difference between the objective function when the matching threshold is 1 and 0. Our sensitivity analyses shed light on the effect of conservativeness in the objective function and allocation decisions as well as the model׳s tractability.