| This research develops a methodology for decision making in organizationally distributed systems. In a distributed system, the decision authorities and information are dispersed in multiple organizations. A distributed model is proposed that is based on the cooperative interaction among organizations and coupling agents to achieve a global goal. Organizations partially share local information with others through interactions such that the lack of complete information of the global system is compensated. Coupling agents resolve conflicting interests among organizations and direct locally optimized solutions to a globally optimized solution. In this approach, no organization or coupling agent have complete information about the entire system. The proposed model is a generic model that can be applied to decision making problems with a variety of mathematical structures. The proposed mechanism is applied to parameter design problems, capacity allocation problems and machine scheduling problems to illustrate the behavior of the proposed methodology in the realm of nonlinear optimizations, linear programs and integer programs respectively. Experimental studies suggest near optimal results in the problem context tested. |
