Deep uncertainty in humanitarian logistics operations: decision-making challenges in responding to large-scale natural disasters

Abstract

Humanitarian logistics operations perform challenging tasks while responding to large-scale natural disasters. Decision makers at different stages of humanitarian operations exploit numerous problem-specific decision-making models or tools. When synchronising the outputs (decisions) from models into a unified solution, the situation becomes critical because of the lack of consensus on objectives and the availability of model alternatives with uncertainty in the models' key parameters and evaluation of the models' alternative outcomes. Thus, the operational environment becomes complex to respond urgently to humanitarian needs and makes the situation deeply uncertain. In this paper, we inspect humanitarian logistics problems and available deep uncertainty approaches to identify the adapting needs in the latter to be applicable to the former. Our research findings indicate that deep uncertainty approaches should incorporate the concept of short-term planning by considering time constraints, bounded process iteration, data transformation technique, handling process failure, and ways of identifying model assumptions.