home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!caen!hellgate.utah.edu!cc.usu.edu!grdg5
- From: grdg5@cc.usu.edu
- Newsgroups: sci.math.num-analysis
- Subject: Decomposing Linear Programming problem?
- Message-ID: <1992Sep4.175601.58642@cc.usu.edu>
- Date: 4 Sep 92 17:56:01 MDT
- Organization: Utah State University
- Lines: 25
-
- Hi there,
-
- I am looking for a solution method to solve a linear programming
- problem having the form:
-
- Minimize c'x + d'y + e'z ( ' denotes transpose )
- x,y,z
- P1
- subject to Ax + By + Cz <= b
- x >=0 , y >=0, z >= 0
-
- where x, y, and z are vectors of design variables of dimensions
- nx, ny, and nz, respectively, and c, d, and e the corresponding
- cost vectors. Since nx, ny, nz, and the number of
- inequality constraints are large, I'm interested in a solution method
- which decomposes problem P1 into small subproblems for
- x, y, and z separately. However, they are interrelated through a master
- problem.
-
- Does anyone run into this kind of problems or have any ideas about such
- solution technique?
-
- Thanks in advance.
- K. Chompooming
- Civil Eng., Utah State U.
-