论文信息 - A Class of Generalized Greedy Algorithms for the Multi-Knapsack Problem

A Class of Generalized Greedy Algorithms for the Multi-Knapsack Problem

A class of generalized greedy algorithms is proposed for the solution of the [lcub]0,1[rcub] multi-knapsack problem. Items are selected according to decreasing ratios of their profit and a weighted sum of their requirement coefficients. The solution obtained depends on the choice of the weights. A geometrical representation of the method is given and the relation to the dual of the linear programming relaxation of multi-knapsack is exploited. We investigate the complexity of computing a set of weights that gives the maximum greedy solution value. Finally, the heuristics are subjected to both a worst-case and a probabilistic performance analysis.

Alexander H. G. Rinnooy Kan | Leen Stougie | Carlo Vercellis

[1] Jan Karel Lenstra,et al. Two Lines Least Squares , 1982 .

[2] George S. Lueker,et al. On the Average Difference between the Solutions to Linear and Integer Knapsack Problems , 1982 .

[3] Gregory Dobson,et al. Worst-Case Analysis of Greedy Heuristics for Integer Programming with Nonnegative Data , 1982, Math. Oper. Res..

[4] L. J. Savage,et al. Three Problems in Rationing Capital , 1955 .

[5] W. Hoeffding. Probability Inequalities for sums of Bounded Random Variables , 1963 .

[6] Rainer Schrader,et al. ON THE EXISTENCE OF FAST APPROXIMATION SCHEMES , 1981 .

[7] Carlo Vercellis,et al. Stochastic on-line knapsack problems , 1995, Math. Program..

[8] A. Frieze,et al. Approximation algorithms for the m-dimensional 0–1 knapsack problem: Worst-case and probabilistic analyses , 1984 .

[9] David S. Johnson,et al. Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[10] Alexander H. G. Rinnooy Kan,et al. A probabilistic analysis of the multiknapsack value function , 2015, Math. Program..