Title :
Lexicographic α-Robust Knapsack Problem : Complexity Results
Author :
Kalaï, Rim ; Vanderpooten, Daniel
Author_Institution :
LAMSADE, Paris-Dauphine Univ., Paris
Abstract :
The knapsack problem is a classical combinatorial problem used to model many industrial situations. Faced with uncertainty on the model parameters, robustness analysis is an appropriate approach to find reliable solutions. The robust knapsack problem was studied in the literature using a max-min criterion. In this paper, we show the drawbacks of this criterion and propose a new robustness approach, called lexicographic α-robustness. We show that the complexity of the lexicographic α-robust problem does not increase compared with the max-min version and present a pseudo-polynomial algorithm in the case of a bounded number of scenarios.
Keywords :
combinatorial mathematics; computational complexity; estimation theory; knapsack problems; minimax techniques; combinatorial problem; industrial situations; lexicographic alpha-robust knapsack problem; max-min criterion; pseudo-polynomial algorithm; Advertising; Computer applications; Costs; Investments; Minimax techniques; Portfolios; Production; Research and development; Robustness; Uncertainty; 0-1 Knapsack problem; Complexity; Lexicographic ¿-robustness; Uncertainty;
Conference_Titel :
Service Systems and Service Management, 2006 International Conference on
Conference_Location :
Troyes
Print_ISBN :
1-4244-0450-9
Electronic_ISBN :
1-4244-0451-7
DOI :
10.1109/ICSSSM.2006.320662