[Previous] [Next]

XPRBminim

Purpose

Minimize the objective function for the active problem.

Synopsis

int XPRBminim(XPRBprob prob, char *flags);  

Arguments

prob  

Reference to a problem.

flags  

Choice of the solution algorithm, which may be one of:

" "	solve the problem using the recommended LP/QP algorithm (MIP problems remain in presolved state);
"d"	solve the problem using the dual simplex algorithm;
"p"	solve the problem using the primal simplex algorithm;
"b"	solve the problem using the Newton barrier algorithm;
"n"	use the network solver (LP only);
"l"	relax all global entities (integer variables etc) in a MIP/MIQP problem and solve it as a LP problem (problem is postsolved);
"g"	solve the problem using the MIP/MIQP algorithm. If a MIP/MIQP problem is solved without this flag, only the initial LP/QP problem will be solved.

Return value

0 if function executed successfully, 1 otherwise.

Example

The following code minimizes the objective function of expl2 using the Newton barrier algorithm.

XPRBprob expl2;
expl2 = XPRBnewprob("example2");
   ...
XPRBminim(expl2, "b");

Further information

This function selects and starts the Xpress-Optimizer solution algorithm. The flags indicating the algorithm choice may be combined where it makes sense, e.g. "dg. If the matrix loaded in the Optimizer does not correspond to the current state of the specified problem it is regenerated automatically prior to the start of the algorithm. Matrix reloading can also be forced by calling XPRBsync before the optimization. Before solving a problem, the objective function must be selected with XPRBsetobj. Note that if you use an incomplete global search you should finish your program with a call to the Optimizer library function XPRSinitglobal in order to remove search tree information that has been stored, or else you may not be able to rerun your program.