Published online by Cambridge University Press: 27 April 2010
In the last decade, several robustness approaches have beendeveloped to deal with uncertainty. In decision problems, andparticularly in location problems, the most used robustnessapproach rely either on maximal cost or on maximal regretcriteria. However, it is well known that these criteria are tooconservative. In this paper, we present a new robustness approach,called lexicographic α-robustness, which compensatesfor the drawbacks of criteria based on the worst case. We applythis approach to the 1-median location problem under uncertaintyon node weights and we give a specific algorithm to determinerobust solutions in the case of a tree. We also show that thisalgorithm can be extended to the case of a general network.