Yet Another Math Programming Consultant

I am a full-time consultant and provide services related to the design, implementation and deployment of mathematical programming, optimization and data-science applications. I also teach courses and workshops. Usually I cannot blog about projects I am doing, but there are many technical notes I'd like to share. Not in the least so I have an easy way to search and find them again myself. You can reach me at erwin@amsterdamoptimization.com.

Monday, August 23, 2010

Long running horizon

›
I am experimenting with a timber model. These models have typically long horizons, which can cause some problems with some solvers. As the m...
2 comments:
Thursday, August 19, 2010

Cplex new version

›
In some cases a new version of a solver can perform worse on some models. Here is a particular unfortunate example, where Cplex 12.1 finds s...
3 comments:
Saturday, August 14, 2010

QP Models with GAMS/Knitro

›
> My large QP portfolio model does not solve with GAMS/KNITRO. Yes, for large problems you will get: Memory estimate for Hessian i...
6 comments:
Wednesday, August 11, 2010

Simplex vs barrier

›
For some models the barrier algorithm is very fast compared to the Simplex method (for an example see http://yetanothermathprogrammingconsul...
2 comments:
Tuesday, August 10, 2010

Gurobi recovery in difficult LP

›
It is interesting to see how Gurobi recovers from numerical trouble in a difficult (dense) LP, when using the primal simplex algorithm: It...
1 comment:
Monday, August 9, 2010

OR in MIP

›
Question: If we have two non-negative variables, X1, X2 >= 0 and we know at least one of them is zero. How shall we linearize the cons...

Solver Foundation CSP: Min/Max

›
>I want to use Min and Max operators to find the Minimum and maximum value for an array. This array has >decision variables invol...
2 comments:
Saturday, July 31, 2010

Cplex log

›
I spent a large amount of time wasting my eyes while staring at solver progress logs. Sometimes they provide some surprises: Starting C...
4 comments:
Tuesday, July 27, 2010

NLP reformulation

›
In a huge NLP model (> 1e5 equations) the following constraint caused problems: REL_P(S,R).. P_L(S,R) =E= P(S,R)**E('elas') ...
Monday, July 26, 2010

Scheduling problem formulation

›
A prototype model was sent to me that was basically a job-shop model. It was formulated with a time index. Unfortunately this model had a la...
2 comments:
‹
›
Home
View web version

About Me

Erwin Kalvelagen
View my complete profile
Powered by Blogger.