Operations Research 2025
Abstract Submission

2345. Software for Verified Optimization of Mixed-Integer Linear Programs

Invited abstract in session WC-9: What's New in Solvers II, stream Software for Operations Research.

Wednesday, 13:30-15:00
Room: H15

Authors (first author is the speaker)

1. Ambros Gleixner
HTW Berlin
2. Leon Eifler
Zuse-Institute Berlin (ZIB)
3. Alexander Hoen
Zuse Institute Berlin
4. Magnus Myreen
Chalmers University
5. Yong Kiam Tan
Institute for Infocomm Research, A*STAR and Nanyang Technological University

Abstract

Standard solver software for mixed-integer linear programming relies on floating-point arithmetic and defines feasibility and optimality of solutions within numerical tolerances. The correctness of its results, even within these tolerances, is subject to roundoff errors. In this talk, we give an overview of numerically exact software for solving mixed-integer linear programs without tolerances and for verifying the correctness of the result. In particular, we highlight the exact functionalities of the newly released version 10 of the open-source MIP solver SCIP.

Keywords

Status: accepted


Back to the list of papers