A Novel Discrete Global-Best Harmony Search Algorithm for Solving 0-1 Knapsack Problems

摘要: In order to better solve discrete 0-1 knapsack problems, a novel global-best harmony search algorithm with binary coding, called DGHS, is proposed. First, an initialization based on greedy mechanism employed improve the initial solution quality in DGHS. Next, we present improvisation process intuitive cognition of improvising new harmony, which best memory (HM) used guide searching direction evolution during consideration, or else randomly chosen from HM and then genetic mutation done some probability phase pitch adjustment. Third, two-phase repair operator infeasible vector further feasible solution. Last, selection scheme is applied decide whether not generated included into HM. The proposed DGHS evaluated twenty problems different scales compared other three metaheuristics literature. experimental results indicate that efficient, effective, robust for solving difficult problems.