MCDM'15 - paper no. 11
MCDM APPLICATIONS OF NEAR OPTIMAL SOLUTIONS IN DYNAMIC PROGRAMMING
Tadeusz Trzaskalik
Abstract:
One of the methods of scalarization of a multi-criteria problem is the application of a quasi-hierarchy, determined by the decision maker. In discrete problems, to apply this method it is necessary to have an algorithm which generates the optimal solution and the consecutive solutions, contained within the tolerance interval determined by the decision maker. This paper presents algorithms generating the consecutive realizations for a multi-stage deterministic decision-making process as well as an algorithm generating the consecutive strategies for a multi-stage stochastic decision-making process. Algorithms using these solutions in a multi-criteria quasi-hierarchical process are also proposed.
Keywords:
multiple objective dynamic programming, quasi-hierarchy, i-th process realization, i-th strategy, optimality equations.
Reference index:
Tadeusz Trzaskalik, (2016), MCDM APPLICATIONS OF NEAR OPTIMAL SOLUTIONS IN DYNAMIC PROGRAMMING, Multiple Criteria Decision Making (10), pp. 166-184
download