An interactive approach to the stochastic multiobjective allocation problem

Abstract

The paper considers the stochastic multiobjective allocation problem. An assumption is made that a particular resource should be allocated to T projects. The specified level of each goal (along with the known probabilities) can be obtained based on the amount of the resources allocated. We will propose an identification strategy procedure that the decision maker can implement. Our technique combines multiobjective dynamic programming and an interactive approach. First, efficient strategies are identified using Bellman's principle applied to the multiobjective problem. Next, a dialog procedure is used to find a solution that is satisfactory for the decision maker. A numerical example is presented to show how the procedure can be applied.

Keywords

• stochastic allocation problem
• multiobjective dynamic programming
• interactive approach
• stochastic dominance