Abstract
Conference Title: 2014 2nd International Symposium on Computational and Business Intelligence (ISCBI) Conference Start Date: 2014, Dec. 7 Conference End Date: 2014, Dec. 8 Conference Location: New Delhi, India Many problems in multi-agent systems can be described as Distributed Constraint Satisfaction Problems (D CSPs), where the goal is to find a set of assignments to variables that satisfies all constraints among agents. However, when real-life application problems are formalized as DCSPs, they are often over-constrained and have no solution that satisfies all constraints. Moreover, the globalization of the economy and democratization of the Internet, boosted by the huge growth in information and communication technologies, have largely contributed to the expansion of numerous distributed architectures. Thus this paper provides a new distributed management and decision support system suitable to these interdependencies and these complex environments. We present a Distributed Optimization under Constraints Basic Relax (DOC-BRelax) as a new framework for dealing with over-constrained situations. We also present a version of this framework called DOC-Max Relax and a new algorithm for solving Distributed Maximal Constraint Satisfaction Problems (DMCSPs).