Analyzing Performance of GA for its Various Parameters While Solving Multiple Knapsack Problem

Abstract

The Knapsack problem is a problem in combinatorial optimization. Multiple Knapsack problem (MKP) are used for solving different domain of engineering and sciences. Multiple knapsack problem has applications in the field of cloud computing, cryptography, etc. Multiple knapsack problem (MKP) is a alternative of 0/1 in this item are positioned in multiple knapsacks having different capacities. In MKP a subset of items is to be placed in different knapsack in such a way the aggregated gain or profit of the chosen individuals are is optimum deprived of overloading each of the knapsacks. The researcher solved the MKP problem by Genetic algorithm. A Genetic Algorithm (GA) is a technique for computing either conditional or unconditional optimization problem using a neutral secretion procedure like biological process. The Genetic Algorithm is applied to some standard instances of the MKP problem. The performance of GA is analysed on its various parameters like population size, cross over rate, mutation rate. The results discover that; GA performs differently on a different value of parameters.

Keywords: Multiple Knapsack Problem (MKP), Genetic Algorithm (GA), Greedy Method Particle Swarm Optimization