[Lemon-devel] Network Simplex

[Pierre Chardaire (CMP)]

You distinction is not common in the OR community, but I do not care that much provided it is made perfectly clear in the documentation.
In any case, I like the lemon library very much.

Note: I am not totally ignorant in this field, as I designed my own network simplex algorithm for non-oriented multi-commodity flow problems:

@article{DBLP:journals/ior/ChardaireL02,
  author    = {Pierre Chardaire and
               Abdel Lisser},
  title     = {Simplex and Interior Point Specialized Algorithms for Solving
               Nonoriented Multicommodity Flow Problems},
  journal   = {Operations Research},
  volume    = {50},
  number    = {2},
  year      = {2002},
  pages     = {260-276},
  ee        = {http://dx.doi.org/10.1287/opre.50.2.260.436},
  bibsource = {DBLP, http://dblp.uni-trier.de}
}

Cheers,

Pierre.

>-----Original Message-----
>From: Alpár Jüttner [mailto:alpar.juttner at gmail.com] On Behalf Of Alpar
>Juttner
>Sent: Thursday, December 06, 2012 2:03 PM
>To: Kovács Péter
>Cc: Pierre Chardaire (CMP); lemon-devel at lemon.cs.elte.hu
>Subject: Re: [Lemon-devel] Network Simplex
>
>On Thu, 2012-12-06 at 13:28 +0100, Kovács Péter wrote:
>> Hi All,
>>
>> I generally agree with Alpar, but the distinction between invalid and
>> infeasible is not so clear for me in this particular case.
>
>I would say, an input is valid if the Hoffman condition works for that.
>