Title :
Cellular competitive decision algorithm for binary knapsack problem
Author :
Xiaohua Xiong ; Aibing Ning
Author_Institution :
Coll. of Comput. & Inf., Shanghai Second Polytech. Univ., Shanghai, China
Abstract :
Classical binary or 0-1 knapsack problem is one of the most widely studied problems in combinatorial optimization. Though the optimization version of this problem is NP-hard, practical solution techniques don´t require optimality. Many different heuristics and approximation algorithms have been used to solve t it. Cellular competitive decision algorithm (CCDA) is a heuristic proposed recently, which is good at solving combinatorial problems, especially those NP-hard problems. One of the most obvious features of CCDA is easy to integrate with the mathematical properties of problem itself. Firstly, mathematical properties of binary knapsack problem are analyzed to reduce the scale of problem. Then a CCDA for binary knapsack problem is presented based on the feature of binary knapsack problem. In addition to validate the effect and efficiency of the algorithm, we conducted many computational investigations to verify the feasibility of CCDA. The results suggest the proposed algorithm is extremely well with regard to both running time and quality of the solutions found.
Keywords :
approximation theory; cellular automata; combinatorial mathematics; computational complexity; knapsack problems; 0-1 knapsack problem; NP-hard problem; approximation algorithm; binary knapsack problem; cellular competitive decision algorithm; combinatorial optimization; Algorithm design and analysis; Approximation algorithms; Automata; Educational institutions; Heuristic algorithms; Optimization; Upper bound; algorithm; cellular automaton; competitiveness function; decision function; knapsack problem;
Conference_Titel :
Natural Computation (ICNC), 2014 10th International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4799-5150-5
DOI :
10.1109/ICNC.2014.6975886