Modeling of imprecise cause & effect relations in managerial systems using fuzzy inference network

Modeling of uncertainty is the main goal of modern scientific view in System Era. The challenge in complex problems such as managerial decision making is to model relations between components which cause uncertainty in real world problem. Dealing with imprecise data in nonlinear environment and in a lack of historical data is very usual in managerial problems where there are no sufficient useful tools to solve them. System Dynamics is a dominant tool that helps policy makers to simulate the behavior of the system avoiding catastrophic decision making resulting from trial & error in real situations. In this paper Fuzzy Inference System, as a powerful tool, is used in a network to simulate the cause and effect relations in complex systems over the time (as a supportive variable). The advantages of introduced model are dealing with imprecise data, employing knowledge of experts and considering hard and soft data in parallel. Also, this model benefits from avoiding simplification assumptions such as linearization. To illustrate the applicability of this model productivity problem of a company is considered as a case study. The obtained results were unexpected but surprisingly they support the company needs. The results are approved by high rank personnel in company and have shown the advantage of using knowledge depository of experts based on real relation in local situation instead of global rules.