AN INTELLIGENT DECISION SUPPORT BASED ON A SUBTRACTIVE CLUSTERING AND FUZZY INFERENCE SYSTEM FOR MULTIOBJECTIVE OPTIMIZATION PROBLEM IN SERIOUS GAME

Abstract

Learning decision making through playing a game is an interesting activity for the decision maker or player. In this paper, a multiobjective optimization problem for economic and emission dispatch in which the player can learn about the tradeoff between fuel cost (economic) and emission problems to achieve optimal decisions is considered. A nonplayer character (NPC) is an entity that is built to provide intelligent decision support for the player. The proposed approach is carried out in two stages for the NPC module: the first stage uses the nondominated sorting genetic algorithm II method to solve the multiobjective optimization problem; this stage produces some optimal solutions. The next stage uses subtractive clustering to cluster optimal solutions; furthermore, these clusters are used to build a fuzzy inference system based on the Mamdani type. In this stage, players can select the best decision offered by the NPC.