Generation techniques for linear programming instances with controllable properties

Simon Bowly, Kate Smith-Miles, Davaatseren Baatar, Hans Mittelmann

Research output: Contribution to journalArticle

Abstract

This paper addresses the problem of generating synthetic test cases for experimentation in linear programming. We propose a method which maps instance generation and instance space search to an alternative encoded space. This allows us to develop a generator for feasible bounded linear programming instances with controllable properties. We show that this method is capable of generating any feasible bounded linear program, and that parameterised generators and search algorithms using this approach generate only feasible bounded instances. Our results demonstrate that controlled generation and instance space search using this method achieves feature diversity more effectively than using a direct representation.

Original languageEnglish (US)
Pages (from-to)389-415
Number of pages27
JournalMathematical Programming Computation
Volume12
Issue number3
DOIs
StatePublished - Sep 1 2020

Keywords

  • Controllable properties
  • Encoded space
  • Instance generation
  • Linear programming

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Software

Fingerprint Dive into the research topics of 'Generation techniques for linear programming instances with controllable properties'. Together they form a unique fingerprint.

  • Cite this