tag:blogger.com,1999:blog-593563533834706486.post4540761170955942245..comments2024-03-28T10:35:10.453-04:00Comments on Yet Another Math Programming Consultant: Cplex messageErwin Kalvelagenhttp://www.blogger.com/profile/09496091402502236997noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-593563533834706486.post-41437166145947623722009-01-29T14:12:00.000-05:002009-01-29T14:12:00.000-05:00Indeed: I can confirm I have not seen this message...Indeed: I can confirm I have not seen this message recently.Erwin Kalvelagenhttps://www.blogger.com/profile/09496091402502236997noreply@blogger.comtag:blogger.com,1999:blog-593563533834706486.post-59658625813881249862009-01-22T12:43:00.000-05:002009-01-22T12:43:00.000-05:00This is a response to a rather old posting by now,...This is a response to a rather old posting by now, but I can tell you that CPLEX 11.1 contains an enhancement to Solution Polishing that makes the condition vanishingly rare. In most cases the error message used to come when a solution provided to CPLEX did not fit the presolved model; for example, perhaps presolve can prove that in any optimal solution, x=0, and yet x=1 is a perfectly feasible (but suboptimal) choice. CPLEX 11.1 now handles this case.<BR/><BR/>John Gregory<BR/>CPLEX Product Manager <BR/>ILOG, an IBM CompanyAnonymousnoreply@blogger.com