Computational Experience with Optimal Value Function and Lagrange Multiplier Sensitivity in NLP
Abstract
Sensitivity analysis in nonlinear programming is examined using the Sequential Unconstrained Minimization Technique. Particular emphasis is placed on the partial derivatives of the Lagrange multipliers and the optimal value function taken with respect to specified problem parameters, the estimation of which is based on recent developments by Armacost and Fiacco. The computational experience complements that previously reported by Armacost and Fiacco. An application of the sensitivity analysis to a large-scale, multi-item inventory model developed for the U.S. Navy is presented.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 10, 1976
- Accession Number
- ADA026374
Entities
People
- Robert L. Armacost
Organizations
- George Washington University