Research On Principal-Agent Problems Of Hospital Management In Uncertain Environment

Principal-agent theory has more than fifty years' history. From the uncertain perspective of information asymmetry and interests inconsistency between principal and agent, and agent's economic behavior results, the theory put its research on enterprise inner problem including agent's incentive problem and agent's risk-sharing issue etc.The dissertation reviews the history of principal-agent theory and concludes the related conception and character in the theory, then gives the basic models of the theory. The principle-agent relationship in hospital is analyzed and the principle-agent models in hospital in uncertain environment are proposed. The relationship between the owner and the manager in the case of incomplete information is analyzed too. Based on the relationships in hospital analyzed, the dissertation models the problem in hospital using principle-agent theory.The dissertation considers principle may use many incentive mechanisms including ration or ratio compensating. Sometimes the principle may make agent share all the risk. In the mechanism of ration compensating, the incentive mechanism has shortcoming which can be compensated by ratio compensating incentive mechanism adopting transfer payoff. And then two-level expected programming model in fuzzy environment satisfying discrete dynamic system is constructed. In the model, the lower level management is perform due to the upper level decision.In the solving models with fuzzy parameters, the dissertation coverts the models into expected ones because it is no meaning of optimize fuzzy variable. It is very difficult to analyze the continuity and the character of concave -convex about the expected models covert so the analytic solution can not be obtained. A hybrid intelligent algorithm which mixed fuzzy simulation and particle swarm optimization is designed to solve the models constructed. In the complete information dynamic incentive models, the constraint level has district dynamic system, it is considered to solve by dynamic programming. Unfortunately, there are fuzzy parameters in the models and it method is not able to solve the model. In order to solve the models, the hybrid intelligent algorithm which mixed fuzzy simulation, dynamic programming and particle swarm optimization is designed.