Proc OPTMODEL : Order of a set | GAMS <--> Optmodel comparison sheet | MINLP solver

[ODENWALD]

Hi :

I have two questions :

1.  Frequently one needs a term as   j * T[j] :  that's no problem if  j is numeric. However if you have { j in ARCS}  and your  ARCS  are  <NY,SF>  or the like you would need something as the  order function in  GAMS  ord(j) :  What is the equivalent in Optmodel ? Add :  When I write  { j in ARCS}  essentially I leave it to SAS to determine the order in which the arcs are processed: how can I recover that order ?

2.  People have to use what they find at their places. So it can occur that one has to translate a model formulation from  GAMS  to  SAS (Optmodel)  ore vice versa. Are you aware of any compilation of a kind of correspondence table between the two systems ?

3.  What is the state of the affairs with a  MINLP solver for Optmodel ??  GAMS  can  access one.

SAS  BLACKBOX -  and  LSO - solvers both say they would perform with  around  100  variables.

Frequently you find yourself with around   600  which can technically go to  ~ 2000  variables  ......

I understand that  Q3  needs time for answering.  Q1  and  Q2  should go more easily.

Kind regards,  ODENWALD .

[RobPratt · Accepted solution]

1. Here's a simple example that creates an ord array to access the set members by position:

proc optmodel; 
   set S = /A B C/; 
   str ord {1..card(S)}; 
   num count init 0; 
   for {i in S} do; 
      count = count + 1; 
      ord[count] = i; 
   end; 
   print ord; 
   print ord[2];
quit;

2. I'm not aware of any such table.

3. We do have long-term plans to provide a generic MINLP solver.  For now, Blackbox is the best choice.  Coming soon, OPTMODEL will provide automated linearization for several common use cases.  Stay tuned!

[RobPratt]

The linearization feature I mentioned is now available.  The idea is that you use the nonlinear expressions directly and then call:

solve linearize;

For several common linearizable use cases, OPTMODEL introduces the new variables and constraints for you under the hood.

[ODENWALD]

Rob :  Thanks for the info.

Is it in  (SAS 9.4, Rev. 940_20w39) or in a more recent one

which is not yet listed ?

Odenwald

[RobPratt]

It is in SAS Optimization in SAS Viya 2020.1.1.

[ODENWALD]

When will it show up in  SAS/OR ?

Odenwald

[RobPratt]

That is yet to be determined.