About ODENWALD

ODENWALD · ‎12-23-2020

When will it show up in SAS/OR ? Odenwald

ODENWALD · ‎12-23-2020

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

ODENWALD · ‎12-08-2020

Hi Steven : Unfortunately I cannot present the code publicly. We are working on linearization , seems to work ; we get solutions that are worth being discussed, albeit very wide bounds. We hope to get forward. I ran the more or less toy examples that come with Blackbox. That worked. I tried 3 real ones and in any case obtained "Failed" in all variants that I tried. I'm far from saying that my cases are the world but SAS publishing more advice on how to effectively use Blackbox could be helpful. Rob Pratt recently remarked that some kind of automatic linearization would be coming soon (soon is my formulation). In the 1990s SAS has written quite some procedures to attack or solve non-linear statistical problems that used an exact or approximate linearization to get a. feasible points and/or b. good starting points for non-linear iterations. It would be great if that strategy could be implemented for Blackbox, too. Odenwald .

ODENWALD · ‎12-07-2020

Hi : Your outline is a bit sketchy. Are a and b dimensions of your array that you seek ? Then you had an MINLP-Problem. Or is your F(.) F(x1, x2a, a,b, x2b, x3, x4) with x2b = 30 fixed and x2b = 100 fixed ? You could Optmodel tell so. You would use the impvar-construction and a solver presumably of type Blackbox or LSO or NLP. Odenwald

ODENWALD · ‎12-07-2020

Hi : I do supervise a student who does an internship in a company and uses the company's SAS System via EG. SAS runs on some server. Main tool is OR- Proc Optmodel. I'm in Home Office and can access SAS, SAS Studio or EG - my EG is running locally. I use EG very sparsely. Difference : In EG I see the SAS Log as usual enfolding as the program (solver) runs. The student says I can see no SAS Log. What had the student to do to see the SAS log when in EG the Optmodel's Solver iterates ?? It's essential to monitor the progress. Kind regards, Odenwald

ODENWALD · ‎12-07-2020

Hi: For the first time in my life I used the new Blackbox Solver on a moderately non-linear problem. I started between 10 and 20 GAs. In any attempt the result was Failed with numbers for Infeasibility which seemed rapidly going in a kind of stalling. Also I could not detect a result of the say 10 global GAs whence I had expected to somehow "see" different partial results from differing global GAs as e.g. a kind of jumping in the values of Infeasibilty. I got not one single feasible solution. It appeared that all the GAs searched in a false region of the search space. Is there a way to throw out the GAs into a wider range of the search space ? If you have some vague idea where a good solution could be how can I use that info as kind of starting values for some variables (which I teach since ages are very important in solving non-linear problems) ? Is there a hidden option to the end to tell the solver please find me just a feasible solution ? I was a bit disappointed when comparing GAMS with a MINLP-solver with Blackbox. Kind regards, Odenwald.

ODENWALD · ‎12-07-2020

Hi : When I work with Optmodel via Enterprise Guide(the only way to work) it is normal that I have solve .... ; print X Y Z ; When it takes for ages to solve a MILP-Problem to optimality in SAS I do : Ctrl + Alt + Break Select not Cancel, but HALT, Confirm and view the results of the print statements as the Solver is halted but the print is processed and I can view the best integer solution found so far. I could NOT perform the same under EGuide as I found several ways to STOP/Cancel but no way to HALT. I simply want to interrupt the solver but see the print results. Can you help ? Regards, Odenwald

ODENWALD · ‎12-02-2020

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 .

ODENWALD · ‎11-12-2020

Hi Rob : Perfect. The question as posed is totally solved. Thank you again. Unfortunately the real data which we are not allowed to share are such that you don't see much, but produce <infeasible>. As we could not find an error, I suspect that these data contain so called positive cycles. These are mentioned some times in the literature but mostly excluded from further discussion (these are the equivalents of negative cycles in shortest path problems). If such effects ever have crossed your path please feel encouraged to comment or recommend a paper. There have been suggested different LP models for the task under consideration (Neos/GAMS ; Hillier-Lieberman ; Domschke (German) ; ) Possibly there could exist further ones that allow to handle that cycle problem. Odenwald

ODENWALD · ‎11-11-2020

Hi Rob : Thanks a lot. There are two questions : 1. How can I print a set of Optmodel ? 2. Are there alternatives to the cross and union sequences ? Could I somehow input a linear list of all pairs of type (predecessor , successor) of activity nodes ? Odenwald

ODENWALD · ‎11-11-2020

Hi : Because I need to add further constraints using Proc CPM was not deemed appropriate. Here https://neos-guide.org/content/project-scheduling-critical-path-method I found an example with GAMS. Being not literate in GAMS I would want to try with Optmodel. The GAMS attempt uses precedence constraints where I don't know the equivalents in Optmodel. I'm looking for a decent model in Optmodel since I need a model roughly 10-20 times larger than the easy example. Here is some part of the link copied in : Did not work !! Please look up the link. I think one would map the nodes A to H onto the numbers 1 to 8 ... Who has good ideas ? Kind regards, Odenwald.

ODENWALD · ‎06-24-2020

I don't even come close to your cited time values. What could be the reasons ?

ODENWALD · ‎06-24-2020

Thanks for your tips. Problem solved. It turned out to be the most straightforward way to use R's mvnormalmixEM package. That did the jobs.

ODENWALD · ‎06-19-2020

Hi : I definitely appreciate your support and I see the problem. Many parts in your program I do understand, some are a black box to me. I ran the program many times with various settings and seeds. I included round(.) in the distance calculation. However the best (smallest) value I could achieve was 631. Should be 603. Did you reach 603 ? Odenwald

ODENWALD · ‎06-18-2020

Yes, the mixture probabilities are needed, too. There seems to be at least some overlap between Procs MBC and GMM , both being Viya only. The basic question was : Is such an approach available via SAS/STAT or /IML, too ? The new question would be : Where can I find a step-by-step recipe on how to solve it via Viya ? Where the point is how to access Viya faster than R: mvnormalmixEM - for an old sas-ler.

Online Status	Offline
Date Last Visited	‎06-01-2021 01:40 PM

Re: Proc OPTMODEL : Order of a set | GAMS <--> Optmodel compariso...

Re: Proc OPTMODEL : Order of a set | GAMS <--> Optmodel compariso...

Re: OPTMODEL : Blackbox Solver

Re: Optimization problem

How to enable SAS Log in EG

OPTMODEL : Blackbox Solver

Enterprise Guide and Optmodel : Stop vs. Halt

Proc OPTMODEL : Order of a set | GAMS <--> Optmodel comparison sh...

Re: Project schedule via LP with Proc Optmodel

Re: Project schedule via LP with Proc Optmodel

Re: Proc OPTMODEL : Order of a set | GAMS <--> Optmodel compariso...

Re: Project schedule via LP with Proc Optmodel

Re: Project schedule via LP with Proc Optmodel

Re: Multivariate Gaussian Mixture Model in SAS/STAT or SAS/IML ???

Re: Proc Optmodel and CVRP : Turning off the decomposition algo ...

Re: windows10 about SAS9.4 ERROR: The Java child process terminated un...

Re: Proc OPTMODEL : Order of a set | GAMS <--> Optmodel compariso...

Re: Proc OPTMODEL : Order of a set | GAMS <--> Optmodel compariso...

Re: OPTMODEL : Blackbox Solver

Re: Optimization problem

How to enable SAS Log in EG

OPTMODEL : Blackbox Solver

Enterprise Guide and Optmodel : Stop vs. Halt

Proc OPTMODEL : Order of a set | GAMS <--> Optmodel comparison sh...

Re: Project schedule via LP with Proc Optmodel

Re: Project schedule via LP with Proc Optmodel

Project schedule via LP with Proc Optmodel

Re: Proc Optmodel and CVRP : Turning off the decomposition algo ...

Re: Multivariate Gaussian Mixture Model in SAS/STAT or SAS/IML ???

Re: Proc Optmodel and CVRP : Turning off the decomposition algo ...

Re: Multivariate Gaussian Mixture Model in SAS/STAT or SAS/IML ???