-
Notifications
You must be signed in to change notification settings - Fork 8
Add Dominguez-Rios et al algorithm #28
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's finish #22 first, and then think about which bits they have in common that we can share code between.
Codecov ReportBase: 94.41% // Head: 95.53% // Increases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## master #28 +/- ##
==========================================
+ Coverage 94.41% 95.53% +1.12%
==========================================
Files 6 7 +1
Lines 519 650 +131
==========================================
+ Hits 490 621 +131
Misses 29 29
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
Merging. We can fix the 32-bit solves (I think it's a HiGHS issue?) and add |
Link for the algorithm [1]: https://doi.org/10.1016/j.ins.2021.02.074
This is an algorithm to solve multi-objective discrete optimization (MODO) problems with 2 or more objectives.
This algorithms maintains a list of regions in the outcome space similar to Kirlik & Sayın [2]. The difference is that regions are categorized according to their so-called 'directions'. For a MODO problem with p-objectives, the regions of the outcome space are represented as
Box
es in p-dimensional space (Kirlik & Sayın maintainedRectangle
s in p-1 dimensional space).This is an "anytime algorithm" meaning that during the solution process the decision maker can stop the process anytime and the resulting pareto set will still be a good representation of the original pareto set. For this reason, I believe
SolutionLimit
can be added as an attribute (and maybe a time limit but I do not know if that is already a part of JuMP. It probably is).Test cases are the same as Kirlik & Sayın.
[1]Miguel Angel Dominguez-Rios, FranciscoChicano, and Enrique Alba. Effective anytime algorithm for multiobjective combinatorial optimization prob- lems. Information Sciences, 565:210–228, 7 2021.
[2] Gokhan Kirlik and Serpil Sayin. A new algorithm for generating all nondominated solutions of multiobjective discrete optimization problems. European Journal of Operational Research, 232:479–488, 2 2014.