Difference between revisions of "Error Messages/9318"

(Created page with '= Solver Error = The selected engine can not solve a problem of this type. Please select another engine. = Cause = No information about this solver error is available. Please…')
 
Line 5: Line 5:
 
= Cause =
 
= Cause =
  
No information about this solver error is available. Please feel free to add content, examples, and models about this error.
+
Analytica Optimizer can make use of several different optimization engines to solve an optimization problem.  The various engines can handle different types of problems.  For example, the "LP/Quadratic" engine can solve linear programs and quadratic programs with linear constraints. That engine, however, cannot be used to solve problems with a quadratic or non-linear constraint, or problems with a non-quadratic (i.e., general non-linear) objective function.
  
 +
You may encounter this error when you explicitly specify the solver engine in the «engine» parameter to [[LpDefine]], [[QpDefine]], [[NlpDefine]] or [[DefineOptimization]].  By specifying the engine explicitly, it forces the optimizer to use that engine for the problem, and if the engine is inappropriate, this error results.
 +
 +
If you get this error and you haven't specified the «engine» parameter explicitly, then Analytica has selected what it thinks is probably the most appropriate engine, but managed to pick one that doesn't match the problem. 
 +
 +
The following table lists the available engines that the problem types that they are capable of solving (QP-LC=Quadratic objective with linear constraints, QP-Cone = Quadratic w/conical constraints, QP-CVX = QP with convex quadratic obj and constraints).
 +
 +
{| border="1" style="text-align:center"
 +
! Engine !! LP !! QP-LC !! QP-Cone !! QP-CVX !! NLP-Smooth !! NLP-NonSmooth
 +
|-
 +
! QP/Quadratic
 +
| Y || Y || N || N || N || N
 +
|-
 +
! SOCP Barrier
 +
| Y || Y || Y || ? || N || N
 +
|-
 +
! GRG Nonlinear
 +
| Y || Y || Y || Y || Y || (N)
 +
|-
 +
! Evolutionary
 +
| Y || Y || Y || Y || Y || Y
 +
|-
 +
! LSLP
 +
| Y || Y || N || N || N || N
 +
|-
 +
! LSSQP
 +
| Y || Y || Y || Y || N || N
 +
|-
 +
! LSGRG
 +
| Y || Y || Y || Y || Y || (N)
 +
|-
 +
! Gurobi
 +
|-
 +
! MOSEK
 +
|-
 +
! Knitro
 +
|-
 +
! OptQuest
 +
|-
 +
! XPress
 +
|}
 +
 +
= Remedies =
 +
 +
To get around this problem, you simply need to explicitly specify the engine to use in the «engine» parameter to [[LpDefine]], [[QpDefine]], [[NlpDefine]] or [[DefineOptimization]].  For example:
 +
 +
[[QpDefine]]( ..., engine:"GRG Nonlinear")
  
 
= Remedies =
 
= Remedies =

Revision as of 18:47, 25 February 2010

Solver Error

The selected engine can not solve a problem of this type. Please select another engine.

Cause

Analytica Optimizer can make use of several different optimization engines to solve an optimization problem. The various engines can handle different types of problems. For example, the "LP/Quadratic" engine can solve linear programs and quadratic programs with linear constraints. That engine, however, cannot be used to solve problems with a quadratic or non-linear constraint, or problems with a non-quadratic (i.e., general non-linear) objective function.

You may encounter this error when you explicitly specify the solver engine in the «engine» parameter to LpDefine, QpDefine, NlpDefine or DefineOptimization. By specifying the engine explicitly, it forces the optimizer to use that engine for the problem, and if the engine is inappropriate, this error results.

If you get this error and you haven't specified the «engine» parameter explicitly, then Analytica has selected what it thinks is probably the most appropriate engine, but managed to pick one that doesn't match the problem.

The following table lists the available engines that the problem types that they are capable of solving (QP-LC=Quadratic objective with linear constraints, QP-Cone = Quadratic w/conical constraints, QP-CVX = QP with convex quadratic obj and constraints).

Engine LP QP-LC QP-Cone QP-CVX NLP-Smooth NLP-NonSmooth
QP/Quadratic Y Y N N N N
SOCP Barrier Y Y Y ? N N
GRG Nonlinear Y Y Y Y Y (N)
Evolutionary Y Y Y Y Y Y
LSLP Y Y N N N N
LSSQP Y Y Y Y N N
LSGRG Y Y Y Y Y (N)
Gurobi
MOSEK
Knitro
OptQuest
XPress

Remedies

To get around this problem, you simply need to explicitly specify the engine to use in the «engine» parameter to LpDefine, QpDefine, NlpDefine or DefineOptimization. For example:

QpDefine( ..., engine:"GRG Nonlinear")

Remedies



You are not allowed to post comments.

Comments
<comments />