Probability Collectives for Solving Truss Structure Problems

1. Abstract The approach of Probability Collectives (PC) in the Collective Intelligence (COIN) framework is one of the emerging Artificial Intelligence approaches dealing with the complex problems in a distributed way. It decomposes the entire system into subsystems and treats them as a group of learning, rational and self interested agents or a Multi-Agent System (MAS). These agents iteratively select their strategies to optimize their individual local goal which also makes the system to achieve the global optimum. The approach of PC has been tested and validated by solving a variety of practical problems in continuous domain. This paper demonstrates the ability of PC solving 2-D space truss structure and 3-D truss structure design problems with discrete as well as continuous variables. The approach is shown to be producing competent and sufficiently robust results. The associated strengths, weaknesses are also discussed. The solution to these problems indicates that the approach of PC can be further efficiently applied to solve a variety of practical/real world problems. 2.