Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments March 30, 2023 erstellt
Description:
We define matroid search problems, in which a decision maker (DM) searches for a set of objects where the set of objects and the collection of admissible sets of objects form a matroid. The DM decides whether to inspect each object to learn its value at some cost or to stop searching. We characterize the optimal search value for matroid search problems and introduce “Greedy Pandora’s rule” as an optimal search strategy generalizing Weitzman (1979). As applications, we examine various matroid search problems, such as searches for multiple alternatives, recruiting problems with capacity constraints, assignment problems, and minimum cost spanning tree problems where the values and costs are uncertain