sci.op-research

On Feb 12, 11:47 am, Christian Plum wrote:
> Hi,
>
> I am looking to choose modelling software and a solver for an
> commercial application. The problem will have some integer variables.
>
> A) The model to be implemented concerns minimization of a resource
> cost and relies on two stochastic components, the prices and the
> consumption. So I am looking for modelling software enabling me to
> formulate a stochastic problem, relying on the software to be able to
> convert it into a deterministic problem, solvable by standard solvers.
>
> B) For other projects I have used scripting modelling software as
> GAMS. For the same project i needed to do a lot of preprocessing of
> the data before giving it to GAMS. I found it really annoying that I
> had to keep the preprocessing in one language (which was actually
> capable of numeric manipulation) and construct the model in another
> (where i could construct the model and match the indexes and sets
> automatically). I would really like to find a way to do both..
>
> So I am trying to find modelling software that satisfies both A and B
> and so far i have established that ILOG Cplex satisfy B through its
> "concert technology" and that Mosel-Xpress satifies A, with its Xpress-
> SP.
>
> So Im' asking, Is that correct? Cant I hope to satisfy both at the
> same time ?
>
> Looking forward to your answers! ;)

Let me return to the original problem which is pre-processing, not
solver management during optimization. I happen to use MPL as my
model manager. But the paradigm I've increasingly followed is
separating data management from model management as much as I can. If
you are into a data base to service the model for a penny, you are in
for a pound. So you might as well leverage it's full capabilities
because it's a lot easier to manipulate data on that platform than it
is with GAMS or MPL.

My models can be intermediate sized, and when I need to support one
with a data base, I use Access. I can generate almost every input
vector in Access using the Query Builder including index sets and pass
those such that MPL merely reads in the data/indexes and passes those
on to the with very little manipulation formulation. And error
checking is much easier this way.

I had a recent conversation with Bjarni Kristjansson from Maximal
(MPL) and he acknowledged that his users have migrated in that
direction. It just makes sense.

SteveM