Genetic Algorithm Approach to Entropy Matrix Goal Game via Fuzzy Programming

Abstract

In this paper the entropy function of the matrix game has been considered as one of the objectives and formulate a new game model, named Entropy Matrix Goal Game Model. Non-linear programming model for each player has been established. The equilibrium condition on matrix game is based on expected pay-off, so this equilibrium condition might be violated by mixed strategies when replications are not allowed. To avoid this inconvenience, we have considered G-goal security strategy. The concept of G-goal Security Strategy which assures the property of security against an opponent's deviation in strategy has been introduced. To solve these models, apply fuzzy programming technique through proposed Genetic Algorithm. A numerical example is included to illustrate the results in this paper.