Sequential linear programming with interior-point methods for large-scale optimization

Eric L. Mulkay, S. S. Rao

Research output: Contribution to journalConference article

Abstract

Mathematical researchers have, over the past decade, developed an efficient class of linear programming solvers known as interior-point methods. Interior-point methods have theoretical and observed computational advantage over simplex methods at solving many large linear programming problems and are immune to degeneracy. Common nonlinear programming methods which work well for small and medium sized problems are unable to solve large-scale problems in a timely fashion. Used in an adaptive sequential linear programming strategy, interior-point methods can be a powerful engineering optimization tool. This work demonstrates the application of an adaptive sequential linear programming algorithm that uses an infeasible primal-dual path-following interior-point algorithm and fuzzy heuristics for the solution of large-scale engineering design optimization problems. Numerical examples demonstrate the superiority of interior-point methods compared to well-known simplex-based linear solver in solving large optimum design problems. Superior performance is shown in both computational time and algorithm ability to handle degenerate problems.

Original languageEnglish (US)
Pages (from-to)2175-2180
Number of pages6
JournalCollection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
Volume3
StatePublished - Jan 1 1998
Externally publishedYes
EventProceedings of the 1998 39th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference and Exhibit and AIAA/ASME/AHS Adaptive Structures Forum. Part 1 (of 4) - Long Beach, CA, USA
Duration: Apr 20 1998Apr 23 1998

ASJC Scopus subject areas

  • Architecture
  • Materials Science(all)
  • Aerospace Engineering
  • Mechanics of Materials
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Sequential linear programming with interior-point methods for large-scale optimization'. Together they form a unique fingerprint.

Cite this