Decision Network Planning Model And Its Optimization In Multiple Decision Spaces

In the process of laying a project plan, how to choose a superior one in several feasible programs is an important decision activity to the project manager. In this kind of decision, one feasible program may has some unique key activities which do not belong to other feasible programs. These unique key activities must be expressed by special decision variables and the project program optimization should be carried out through in multiple decision spaces.Decision network planning is one of the best in project program optimization methods. But the model of decision network planning can only experss the quantity differences among the feasible programs and the decision variables are the same in every feasible programs. So it is difficult to express the differences of decision spaces among the feasible programs. To solve the problem, extended model and optimization study of decision network planning is carried out as follows:(1) Express method of decision network planning in in multiple decision spacesWhen decision spaces are different unique variables and rules of the feasible programs should be definitely expressed in decision network planning to the decision makers. So the original model is extended and the decision unit structure which can express several decision spaces is design. Then the modeling method, which combined with scenario analysis method and decision network planning, is put forward.(2) Optimization methods of decision network planning in multiple decision spacesIn different decision spaces, the properties and their states are often essentially different which should be considered by the decision maker. So modern decision theorys,which are combined with the decision unit structure, are used in the optimization of the decision network planning. And the optimization method studies are as follows:Firstly when decision spaces are different the properties and its measurement of feasible programs are often essentially different. So the consolidated method being combined with objective property and the subjective preference is brought forward. And a multiple objectives optimization method is given with ideal point method and multiattribute utility function theory.Secondly in different decision spaces the sample spaces are often essentially different. So a decision unit structure with several sample spaces is put forward. And a random optimization method, which considers expected cost and risk, is given. (3) Case studyA feasible program optimization study of Chemical equipment multimodal transportation project is put forward to validate the model and optimization method.Actually this paper, which is combined with the decision network planning and modern decision theory, mainly focuses on the extened study in expression and optimization of project plan decision. From the effect of the case study, the expression and optimization method, which can optimize the feasible programs in different decision spaces, is closer related to the practice. The accomplished researches of the paper, which obvious have theorical meaning and application value, can provide the theorical and technical support for the project plan decision.