How Near is a Near-Optimal Solution: Confidence Limits for the Global Optimum.

Abstract

For many optimization problems approximate or near-optimal solutions are the only practical solutions available. This paper identifies and compares some procedures which use independent near-optimal solutions to determine a confidence limit such that, with a user specified confidence, the global optimum is between the best near-optimal solution and the confidence limit. A computer implementation of the procedures is available from the authors. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1980
Accession Number
ADA089108

Entities

People

  • Howard M. Monroe
  • Robert L. Sielken Jr.

Organizations

  • Texas A&M University

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Weapons Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Business Administration
  • Computers
  • Confidence Limits
  • Data Science
  • Information Processing
  • Information Science
  • Knowledge Management
  • Mathematical Programming
  • Military Research
  • Network Science
  • Operations Research
  • Optimization
  • Order Statistics
  • Quadratic Programming
  • Statistics
  • Systems Engineering

Fields of Study

  • Mathematics

Readers

  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)
  • Statistical inference.