A comparative study on uncooperative search models in survivor search and rescue

Resource allocation is a challenging topic in the search and rescue in post-disaster process, especially in situations a central management system is not available. Previous studies have well demonstrated the need to improve the search and rescue decision-making in these situations. However, a challenge that current models are facing is how to predict the search cost and efficiency considering the random distribution of survivors with little information or coordination. Based on the discussion on cooperative and uncooperative search and rescue, this paper makes a first attempt to compare the performance of both macroscopic and microscopic search models in search and rescue process, where search and rescue teams are supposed to operate independently, which means in an uncooperative way. The paper further uses the Lévy search in microscopic simulation models, comparing normal random search. The numerical example shows that the macromodel is easy to compute with less demanding data, but its output is more optimistic than that of micromodels, which highlights the need of simulation methods in search and rescue management. Besides, Lévy search is proved to be more efficient than normal random search, especially when the distribution of survivors is sparse and nonuniform; this is mainly because its efficient search area is larger. The findings provide important information for selecting effective analytical models in search and rescue management to save survivors.