Optimization and parameter settings

Functions
int	lp_optimize (LinearProgram *lp)
	Optimizes your Mixed Integer Program. More...
int	lp_optimize_as_continuous (LinearProgram *lp)
	optimizes only the linear programming relaxation of your MIP More...
double	lp_obj_value (LinearProgram *lp)
	objective value of your optimization More...
double	lp_best_bound (LinearProgram *lp)
	returns the best dual bound found during the search More...
double *	lp_x (LinearProgram *lp)
	returns the vector of solution values More...
double *	lp_row_slack (LinearProgram *lp)
double *	lp_row_price (LinearProgram *lp)
double *	lp_reduced_cost (LinearProgram *lp)
	reduced cost for columns - only available when solving continous models More...
int	lp_num_saved_sols (LinearProgram *lp)
	number of solutions stored in the solution pool More...
double	lp_saved_sol_obj (LinearProgram *lp, int isol)
	objective value for the isol-th solution of the solution pool More...
double *	lp_saved_sol_x (LinearProgram *lp, int isol)
	objective value for the isol-th solution of the solution pool More...

Detailed Description

These are the functions that can be used to optimize your model and control some solver configuration.

Function Documentation

lp_best_bound()

double lp_best_bound

(

LinearProgram *

lp

)

returns the best dual bound found during the search

If your Mixed Integer Optimization concluded without finding the optimal solution, you can retrieve the best dual bound (lower bound in minimization), which is a valid estimate of the best possible value for the optimal solution.

Parameters

lp	the (integer) linear program

Returns

the dual bound (lower bound if minimization)

lp_num_saved_sols()

int lp_num_saved_sols

(

LinearProgram *

lp

)

number of solutions stored in the solution pool

Parameters

lp	the (integer) linear program

Returns

number of solutions in the solution pool

lp_obj_value()

double lp_obj_value

(

LinearProgram *

lp

)

objective value of your optimization

Parameters

lp	the (integer) linear program

Returns

objective value found

lp_optimize()

int lp_optimize

(

LinearProgram *

lp

)

Optimizes your Mixed Integer Program.

Optimizes your Mixed Integer Program. Returns the problem status, which can be:

0 : LP_OPTIMAL : optimal solution found
1 : LP_INFEASIBLE : the problem is infeasible
2 : LP_UNBOUNDED : the problem is unbounded
3 : LP_FEASIBLE : a feasible solution was found, but its optimality was not proved
4 : LP_INTINFEASIBLE : the lp relaxation is feasible but no integer feasible solution exists
5 : LP_NO_SOL_FOUND : optimization concluded without finding any feasible solution
6 : LP_ERROR : the solver reported an error

Parameters

lp	the (integer) linear program

lp_optimize_as_continuous()

int lp_optimize_as_continuous

(

LinearProgram *

lp

)

optimizes only the linear programming relaxation of your MIP

Optimizes your Mixed Integer Program relaxing integrality constraints (if any). Returns the problem status, which can be:

0 : LP_OPTIMAL : optimal solution found
1 : LP_INFEASIBLE : the problem is infeasible
2 : LP_UNBOUNDED : the problem is unbounded
3 : LP_FEASIBLE : a feasible solution was found, but its optimality was not proved
4 : LP_INTINFEASIBLE : the lp relaxation is feasible but no integer feasible solution exists
5 : LP_NO_SOL_FOUND : optimization concluded without finding any feasible solution
6 : LP_ERROR : the solver reported an error

Parameters

lp	the (integer) linear program

lp_reduced_cost()

double* lp_reduced_cost

(

LinearProgram *

lp

)

reduced cost for columns - only available when solving continous models

Parameters

lp	the (integer) linear program

Returns

vector with reduced costs

lp_saved_sol_obj()

double lp_saved_sol_obj	(	LinearProgram *	lp,
		int	isol
	)

objective value for the isol-th solution of the solution pool

Parameters

lp	the (integer) linear program
isol	the solution index

Returns

objective value of the i-th solution from the solution pool

lp_saved_sol_x()

double* lp_saved_sol_x	(	LinearProgram *	lp,
		int	isol
	)

objective value for the isol-th solution of the solution pool

Parameters

lp	the (integer) linear program
isol	the solution index

Returns

solution vector of the i-th solution from the solution pool

lp_x()

double* lp_x

(

LinearProgram *

lp

)

returns the vector of solution values

Parameters

lp	the (integer) linear program

Returns

solution vector