.TH glpsol "1" "June 2013" "glpsol: GLPK LP/MIP Solver, v4.51" "User Commands"
.SH NAME
glpsol \- large-scale linear and mixed integer programming solver
.SH SYNOPSIS
.B glpsol
[\fIoptions\fR...] \fIfilename\fR
.SH DESCRIPTION
.SS "General options:"
.TP
\fB\-\-mps\fR
read LP/MIP problem in fixed MPS format
.TP
\fB\-\-freemps\fR
read LP/MIP problem in free MPS format (default)
.TP
\fB\-\-lp\fR
read LP/MIP problem in CPLEX LP format
.TP
\fB\-\-glp\fR
read LP/MIP problem in GLPK format
.TP
\fB\-\-math\fR
read LP/MIP model written in GNU MathProg modeling
language
.TP
\fB\-m\fR filename, \fB\-\-model\fR filename
read model section and optional data section from
filename (same as \fB\-\-math\fR)
.TP
\fB\-d\fR filename, \fB\-\-data\fR filename
read data section from filename (for \fB\-\-math\fR only);
if model file also has data section, it is ignored
.TP
\fB\-y\fR filename, \fB\-\-display\fR filename
send display output to filename (for \fB\-\-math\fR only);
by default the output is sent to terminal
.TP
\fB\-\-seed\fR value
initialize pseudo\-random number generator used in
MathProg model with specified seed (any integer);
if seed value is ?, some random seed will be used
.TP
\fB\-\-mincost\fR
read min\-cost flow problem in DIMACS format
.TP
\fB\-\-maxflow\fR
read maximum flow problem in DIMACS format
.TP
\fB\-\-cnf\fR
read CNF\-SAT problem in DIMACS format
.TP
\fB\-\-simplex\fR
use simplex method (default)
.TP
\fB\-\-interior\fR
use interior point method (LP only)
.TP
\fB\-r\fR filename, \fB\-\-read\fR filename
read solution from filename rather to find it with
the solver
.TP
\fB\-\-min\fR
minimization
.TP
\fB\-\-max\fR
maximization
.TP
\fB\-\-scale\fR
scale problem (default)
.TP
\fB\-\-noscale\fR
do not scale problem
.TP
\fB\-o\fR filename, \fB\-\-output\fR filename
write solution to filename in printable format
.TP
\fB\-w\fR filename, \fB\-\-write\fR filename
write solution to filename in plain text format
.TP
\fB\-\-ranges\fR filename
write sensitivity analysis report to filename in
printable format (simplex only)
.TP
\fB\-\-tmlim\fR nnn
limit solution time to nnn seconds
.TP
\fB\-\-memlim\fR nnn
limit available memory to nnn megabytes
.TP
\fB\-\-check\fR
do not solve problem, check input data only
.TP
\fB\-\-name\fR probname
change problem name to probname
.TP
\fB\-\-wmps\fR filename
write problem to filename in fixed MPS format
.TP
\fB\-\-wfreemps\fR filename
write problem to filename in free MPS format
.TP
\fB\-\-wlp\fR filename
write problem to filename in CPLEX LP format
.TP
\fB\-\-wglp\fR filename
write problem to filename in GLPK format
.TP
\fB\-\-wcnf\fR filename
write problem to filename in DIMACS CNF\-SAT format
.TP
\fB\-\-log\fR filename
write copy of terminal output to filename
.TP
\fB\-h\fR, \fB\-\-help\fR
display this help information and exit
.TP
\fB\-v\fR, \fB\-\-version\fR
display program version and exit
.SS "LP basis factorization options:"
.TP
\fB\-\-luf\fR
LU + Forrest\-Tomlin update
(faster, less stable; default)
.TP
\fB\-\-cbg\fR
LU + Schur complement + Bartels\-Golub update
(slower, more stable)
.TP
\fB\-\-cgr\fR
LU + Schur complement + Givens rotation update
(slower, more stable)
.SS "Options specific to simplex solver:"
.TP
\fB\-\-primal\fR
use primal simplex (default)
.TP
\fB\-\-dual\fR
use dual simplex
.TP
\fB\-\-std\fR
use standard initial basis of all slacks
.TP
\fB\-\-adv\fR
use advanced initial basis (default)
.TP
\fB\-\-bib\fR
use Bixby's initial basis
.TP
\fB\-\-ini\fR filename
use as initial basis previously saved with \fB\-w\fR
(disables LP presolver)
.TP
\fB\-\-steep\fR
use steepest edge technique (default)
.TP
\fB\-\-nosteep\fR
use standard "textbook" pricing
.TP
\fB\-\-relax\fR
use Harris' two\-pass ratio test (default)
.TP
\fB\-\-norelax\fR
use standard "textbook" ratio test
.TP
\fB\-\-presol\fR
use presolver (default; assumes \fB\-\-scale\fR and \fB\-\-adv\fR)
.TP
\fB\-\-nopresol\fR
do not use presolver
.TP
\fB\-\-exact\fR
use simplex method based on exact arithmetic
.TP
\fB\-\-xcheck\fR
check final basis using exact arithmetic
.SS "Options specific to interior-point solver:"
.TP
\fB\-\-nord\fR
use natural (original) ordering
.TP
\fB\-\-qmd\fR
use quotient minimum degree ordering
.TP
\fB\-\-amd\fR
use approximate minimum degree ordering (default)
.TP
\fB\-\-symamd\fR
use approximate minimum degree ordering
.SS "Options specific to MIP solver:"
.TP
\fB\-\-nomip\fR
consider all integer variables as continuous
(allows solving MIP as pure LP)
.TP
\fB\-\-first\fR
branch on first integer variable
.TP
\fB\-\-last\fR
branch on last integer variable
.TP
\fB\-\-mostf\fR
branch on most fractional variable
.TP
\fB\-\-drtom\fR
branch using heuristic by Driebeck and Tomlin
(default)
.TP
\fB\-\-pcost\fR
branch using hybrid pseudocost heuristic (may be
useful for hard instances)
.TP
\fB\-\-dfs\fR
backtrack using depth first search
.TP
\fB\-\-bfs\fR
backtrack using breadth first search
.TP
\fB\-\-bestp\fR
backtrack using the best projection heuristic
.TP
\fB\-\-bestb\fR
backtrack using node with best local bound
(default)
.TP
\fB\-\-intopt\fR
use MIP presolver (default)
.TP
\fB\-\-nointopt\fR
do not use MIP presolver
.TP
\fB\-\-binarize\fR
replace general integer variables by binary ones
(assumes \fB\-\-intopt\fR)
.TP
\fB\-\-fpump\fR
apply feasibility pump heuristic
.TP
\fB\-\-proxy\fR [nnn]
apply proximity search heuristic (nnn is time limit in seconds; default is 60)
.TP
\fB\-\-proxy\fR
apply proximity search heuristic
.TP
\fB\-\-gomory\fR
generate Gomory's mixed integer cuts
.TP
\fB\-\-mir\fR
generate MIR (mixed integer rounding) cuts
.TP
\fB\-\-cover\fR
generate mixed cover cuts
.TP
\fB\-\-clique\fR
generate clique cuts
.TP
\fB\-\-cuts\fR
generate all cuts above
.TP
\fB\-\-mipgap\fR tol
set relative mip gap tolerance to tol
.TP
\fB\-\-minisat\fR
translate integer feasibility problem to CNF\-SAT
and solve it with MiniSat solver
.TP
\fB\-\-objbnd\fR bound
add inequality obj <= bound (minimization) or
obj >= bound (maximization) to integer feasibility
problem (assumes \fB\-\-minisat\fR)
.SH "SEE ALSO"
.TP
For description of the MPS and CPLEX LP formats see the Reference Manual at /usr/share/doc/glpk-doc/glpk.pdf
.PP
For description of the modeling language see "GLPK: Modeling Language
GNU MathProg" at /usr/share/doc/glpk-doc/gmpl.pdf
.PP
Please report bugs to <bug\-glpk@gnu.org>.
.SH COPYRIGHT
Copyright \(co 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008,
2009, 2010, 2011, 2013 Andrew Makhorin, Department for Applied
Informatics, Moscow Aviation Institute, Moscow, Russia.
E\-mail: <mao@gnu.org>.
.PP
This program has ABSOLUTELY NO WARRANTY.
.PP
This program is free software; you may re\-distribute it under the terms
of the GNU General Public License version 3 or later.