Thrifty algorithms for multistage robust optimization

Abstract

We consider a class of multi-stage robust covering problems, where additional information is revealed about the problem instance in each stage, but the cost of taking actions increases. The dilemma for the decision-maker is whether to wait for additional information and risk the inflation, or to take early actions to hedge against rising costs. We study the "k-robust" uncertainty model: in each stage i = 0, 1,..., T, the algorithm is shown some subset of size ki that completely contains the eventual demands to be covered; here k1 > k2 > ⋯ > kT which ensures increasing information over time. The goal is to minimize the cost incurred in the worst-case possible sequence of revelations. For the multistage k-robust set cover problem, we give an O(logm + log n)-approximation algorithm, nearly matching the Ω(log n + log m/log log m) hardness of approximation [4] even for T = 2 stages. Moreover, our algorithm has a useful "thrifty" property: it takes actions on just two stages. We show similar thrifty algorithms for multi-stage k-robust Steiner tree, Steiner forest, and minimum-cut. For these problems our approximation guarantees are O(min{ T, log n, log λmax}), where λmax is the maximum inflation over all the stages. We conjecture that these problems also admit O(1)-approximate thrifty algorithms. © 2013 Springer-Verlag.