PolySCIP is a solver for multi-criteria integer programming and multi-criteria linear programming. In other words, it aims at solving optimization problems of the form:

min/max c 1 x c k x s.t. Axb , x n n

where k 2 , A m × n , b m .

Bi-criteria integer program.
Image of a bi-criteria integer program (where objectives are maximized) with non-dominated points in blue.

Bi-criteria linear program.
Image of a bi-criteria linear program (where objectives are maximized) with non-dominated points in blue.

The name PolySCIP is composed of Poly (from the Greek πολύς meaning "many") and SCIP. The current version of PolySCIP is able to compute supported non-dominated vertices for problems with an arbitrary number of objectives and the entire set of non-dominated points for bi-criteria and tri-criteria integer programs. The file format of PolySCIP is based on the MPS file format.


13/Nov/2015 Website launched.
29/Feb/2016 SCIP Version 3.2.1 with PolySCIP version 1.0 released.
09/Mar/2017 SCIP Version 4.0 with PolySCIP version 2.0 released.
27/May/2017 Visualisation tool PolyNondom available.


The current version of PolySCIP is 2.0 (the development is inactive at the moment). As part of SCIP its source code resides in the directory 'applications/PolySCIP'. You can download the source code via the SCIP website.

A description of features and improvements of PolySCIP 2.0 can be found in section 7.2 of this technical report.


(See also the corresponding INSTALL file in the PolySCIP directory.)

  1. Build SCIP: see the corresponding INSTALL file in the SCIP directory
  2. Build PolySCIP: change into the PolySCIP directory and execute make on the command line
    • if SCIP was built with make [options], then run make [options] with the same options in the PolySCIP directory
  3. Build PolySCIP documentation:
    • Run make doc to build doxygen documentation in 'doc/html'
    • Run cd doc; pdflatex userguide.tex to compile the user guide

How To Cite

Please include a reference if you use PolySCIP for your work:
R. Borndörfer, S. Schenker, M. Skutella, T. Strunk: PolySCIP.
Mathematical Software - Proceedings of ICMS 2016, G.-M. Greuel, T. Koch, P. Paule, A. Sommese (Eds.),
Lecture Notes in Computer Science Vol. 9725, ISBN: 978-3-319-42431-6

User Guide

For more details about the usage, file format of PolySCIP and an easy way to generate .mop problem files containing (your) mathematical programs see the user guide.


Main developer Contributor
Sebastian Schenker Timo Strunk


PolySCIP is part of SCIP and distributed under the ZIB Academic License. You are allowed to retrieve (Poly)SCIP as a member of a non-commercial and academic institution. If you want to use PolySCIP, but you do not comply with the above criteria, please contact me.


If you find any bugs, please send a description.

Problem Library

MOPLIB (short for Multi-Objective Problem LIBrary) is a collection of multi-objective optimization problems. PolySCIP supports the following problem classes: molp, mobp, moip, (momip)

Related Links

If you develop a solver for multi-criteria optimization problems, please let me know.

Bensolve - a vector linear program solver
inner - a multi-objective linear program solver
MOPS - a solver for non-linear multiobjective optimization problems


The development of PolySCIP started in the project A5 Multicriteria Optimisation within the Collaborative Research Center 1026 Sustainable Manufacturing - Shaping Global Value Creation.

Sustainable Manufacturing

Imprint and privacy statement

© 2024 by Zuse Institute Berlin (ZIB). For the imprint and privacy statement we refer to the Imprint of ZIB.