tag:blogger.com,1999:blog-593563533834706486.post7987406120818413481..comments2024-03-09T06:47:13.003-05:00Comments on Yet Another Math Programming Consultant: How to model y=min(x1,x2) Erwin Kalvelagenhttp://www.blogger.com/profile/09496091402502236997noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-593563533834706486.post-72696095065625979812020-02-13T15:41:57.274-05:002020-02-13T15:41:57.274-05:00Thanks. Slightly better than right example, wrong ...Thanks. Slightly better than right example, wrong idea, but I better fix this.Erwin Kalvelagenhttps://www.blogger.com/profile/09496091402502236997noreply@blogger.comtag:blogger.com,1999:blog-593563533834706486.post-64103336160785265392020-02-13T12:44:36.743-05:002020-02-13T12:44:36.743-05:00Your small example model is a bit too simple. Rear...Your small example model is a bit too simple. Rearranging terms, the constraint can be written as max(x1+x2,2x1) = 5. Because of the objective function, you can replace "=" with "<=" (rearranging terms was not necessary, but makes it easier to see that you can turn the equality into an inequality). Since the constraint is convex, you can do the reformulation with the auxiliary variable y, without adding an extra penalty to the objective function (alpha=0 gives the optimal solution).Anonymousnoreply@blogger.com