Robust continuous linear programs (Q2228355)

The robust extension of a linear continuous program (CLP) is reformulated as CLP. The main line of the paper is similar to [\textit{A. Ghate}, Optim. Lett. 10, No. 4, 847--863 (2016; Zbl 1367.90119)]. The satisfaction of the constraints for all uncertainties is represented as an additional finite linear programming problem. The CLP and the finite LP are combined using duality. They get then the reformulated CLP in original and dual variables. Thus the dimension of the reformulated CLP is increased w.r.t. the original problem but all methods for CLP can now be applied for robust CLP too. In the last section upper bounds of the probability for violating the constraints of the original CLP by a solution of the robust CLP are derived. In the contrary to the very structured former paper [loc. cit.], beside the Lemma of the existence of a solution of the CLP, all results must be painfully found by the reader in the running context. Nothing of their main results is explicitly formulated as theorem, proposition or lemma. Thus it is difficult to cite the nice results.