On the Complexity of Calculating Sensitivity Parameters in Boolean Programming Problems
- 49 Downloads
This article shows that, for NP-hard problems, the calculation of even the stability ball of radius 1 for an optimal solution is computationally intensive (i.e., a suitable polynomial algorithm is absent when P ≠ NP). In using greedy algorithms for solving the set covering problem (knapsack problem) with the stability radius r = O(1) , there are polynomial algorithms for calculating the stability ball of radius r for an ln m-approximate solution (1-approximate solution).
Keywordscomplexity of sensitivity analysis stability radius of a problem stability ball of radius r for an ε-approximate problem solution
Unable to display preview. Download preview PDF.
- 1.D. Fernandez-Baca and B. Venkatachalam, “Sensitivity analysis in combinatorial optimization,” in: T. Gonzalez (ed.), Handbook of Approximation Algorithms and Metaheuristics, Chapman&Hall/CRC Computer and Information Science Series, Boca Raton (2007), pp. 30.-1–30.-29.Google Scholar
- 5.I. V. Sergienko, Mathematical Models and Methods for Solving Discrete Optimization Problems [in Russian], Naukova Dumka, Kyiv (1985).Google Scholar
- 6.I. V. Sergienko and N. V. Filonenko, “Solution of some stability problems in integer linear programming,” Dop. AN URSR, Ser. A, No. 6, 79–82 (1982).Google Scholar
- 14.N. N. Kuzyurin and S. A. Fomin, Efficient Algorithms and Complexity of Computations [in Russian], Moscow Institute of Physics and Technology, Moscow (2008).Google Scholar