My Account Log in

1 option

Finding a Dual Feasible Solution to an LP with M Equalities in (l&M) Dual Iterations / Vinay Dharmadhikari.

NBER Working papers Available online

View online
Format:
Book
Author/Creator:
Dharmadhikari, Vinay.
Contributor:
National Bureau of Economic Research.
Series:
Working Paper Series (National Bureau of Economic Research) no. w0100.
NBER working paper series no. w0100
Language:
English
Subjects (All):
Economics.
Economics--Periodicals.
Physical Description:
1 online resource: illustrations (black and white);
Other Title:
Finding a Dual Feasible Solution to an LP with M Equalities in
Place of Publication:
Cambridge, Mass. National Bureau of Economic Research 1975.
Cambridge, Massachusetts : National Bureau of Economic Research, 1975.
Summary:
Lemke's dual-simplex method of linear programming is usually considered inferior to the primal simplex method for any general linear programming problems. One reason given is the difficulty of finding a starting dual-feasible basis. In this paper, a new starting technique is presented, which finds a dual-feasible basis in a single dual-simplex pivot for LP's with no equality constraints, and in (l+m3 ) pivots for LP'S with m3 equality constraints irrespective of the number of inequality constraints. The technique is illustrated on a small example problem. The performance, in terms of the number of pivots to optimality, of the dual-simplex with the new starting technique on 100 medium sized problems is reported and compared with that of the primal simplex. Finally, how the dual-simplex with the new starting technique can be efficiently implemented is briefly discussed.
Notes:
Print version record
August 1975.

The Penn Libraries is committed to describing library materials using current, accurate, and responsible language. If you discover outdated or inaccurate language, please fill out this feedback form to report it and suggest alternative language.

My Account

Shelf Request an item Bookmarks Fines and fees Settings

Guides

Using the Library Catalog Using Articles+ Library Account