Practically, one of the most critical tasks after a nature catastrophe is to organize and execute humanitarian relief operations effectively and efficiently. The humanitarian relief operation is to provide essential daily living supplies to people who are isolated inside the disaster affected areas. However, due to the sudden eruption of large demands, this task becomes challenging for logistics planning authorities. Developing an effective and efficient disaster relief plan is the main motivation in this research. For this purpose, two critical item supply strategies, the centralized-supply and the distributed-supply delivery strategies, are investigated.

In the centralized-supply delivery strategy, a single distribution center (or central depot) and multiple known demand points are considered. Supplies are delivered to demand points directly from the central depot, and all vehicles will return to the central depot after serving demand points. The focus of this part of the study is on modeling a new tour-based logistics formulation that considers delivery priorities, and also on introducing novel heuristic algorithms to manage tours in the model, which play a pivotal role in whether the new formulation can be solved efficiently.

On the other hand, in the distributed-supply delivery strategy, a number of temporary depots are located in the disaster-affected area as intermediate agencies, and, therefore, demand points can be served either by the central depot directly or by temporary depots locally, depending on who can fulfill demand more efficiently. The focus of this part of the study is to propose a framework to locate temporary depots and allocate demand points to exactly one depot appropriately, and to apply formulations and algorithms from the first part of the research (centralized-supply delivery strategy) in each selected temporary depot to execute the humanitarian logistics operation such that the overall performance of the disaster relief operation can be optimized.

In addition to theoretical development, real-world case studies are investigated in this research to gain insights into the practicability of proposed models and solution methodologies. The risk assessment software HAZUS-MH, provided by FEMA, is used, and a historical earthquake disaster in Northridge, California is employed as a scenario. Various analyses are conducted to provide a more comprehensive understanding of how to organize effective and efficient humanitarian logistics planning in a disaster relief operation.