Year: 1977
Type: book
Abstract: In this paper a finite-step method for solving dynamic linear programming (DLP) problems is described. The method suggested is a natural and straightforward extension of one of the most effective static LP methods -- the simplex method -- for DLP. A new concept -- a set of local bases -- for each time step is introduced, thus enabling considerable reduction of the requirements to computer core memory and CPU time. In the proposed method the system of T local m x m bases is introduced and the basic simplex procedures (selection of vectors to be removed from and to be introduced into the basis, pricing procedure, transformation of bases) as applied to this system of T local bases are described. Evaluation of possibilities of the method and its connection with compact inverse LP method are discussed.
Authors: A.I. Propoi, V. E. Krivonozhko
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2526626480