Painted network flows with weighted divergence
Is Version Of
The theory of network flows deals with problems which can be represented as networks and like linear programming provides a general framework for formulating and solving a considerable number of optimization problems. Many network problems can be recast as linear programming problems and vice-versa. A way to treat multistage processes possessing certain invariant aspects is dynamic programming. Certain of these can be recast as network problems;This dissertation is a discussion and detailed description of several algorithms designed to solve two classes of special network problems. One consists of network problems where the flow and the divergence are weighted. And the second class consists of network problems where the potential is weighted. Further, we look at a dynamic programming problem discussed and treated by Kulkarni (1981) and Bechhofer (1985) and formulate it as a network problem which is a special linear optimal distribution problem.