A CGS-MSM PGA Based on Multi-Agent and its Application in Solving 0-1 Knapsack Problem

0-1 knapsack problem is a typical optimization difficult problem. At present, there are a lot of methods to solve this problem, but with the increase of the scale of this problem, among them, the genetic algorithm is an important branch, it is quite active to ask the research of solving 0-1 knapsack on the basis of the genetic algorithm. This paper combines multi-agent theory and CGS-MSM PGA (coarse grain size-master slaver model parallel genetic algorithm) together, propose one improvement CGS-MSM PGA based on multi-agent, then use this PGA to solve 0-1 knapsack problem. This PGA is made up of many agent, solves the 0-1 knapsack problem by the coordination between many agents inside the multi-agent system. The introduction of multi-agent theory, make the master course and slave course of CGS-MSM PGA to be made of agent, so the ability of communication and coordination raise greatly, thus overcome the shortcoming of original CGS-MSM PGA; And comparing with other methods solved 0-1 knapsack problem, the method of this paper not only has the fast computational speed and the high precision, but also can get more optimal solving than other algorithms.