An ε-sensitivity analysis in the primal–dual interior point method
✍ Scribed by Woo-Je Kim; Chan-Kyoo Park; Soondal Park
- Publisher
- Elsevier Science
- Year
- 1999
- Tongue
- English
- Weight
- 124 KB
- Volume
- 116
- Category
- Article
- ISSN
- 0377-2217
No coin nor oath required. For personal study only.
✦ Synopsis
This paper presents a method of sensitivity analysis on the cost coecients and the right-hand sides for most variants of the primal±dual interior point method. We ®rst de®ne an e-optimal solution to describe the characteristics of the ®nal solution obtained by the primal±dual interior point method. Then an e-sensitivity analysis is de®ned to determine the characteristic region where the ®nal solution remains the e-optimal solution as a cost coecient or a right-hand side changes. To develop the method of e-sensitivity analysis, we ®rst derive the expressions for the ®nal solution from data which are commonly maintained in most variants of the primal±dual interior point method. Then we extract the characteristic regions on the cost coecients and the right-hand sides by manipulating the mathematical expressions for the ®nal solution. Finally, we show that in the nondegenerate case, the characteristic regions obtained by e-sensitivity analysis are convergent to those obtained by sensitivity analysis in the simplex algorithm.
📜 SIMILAR VOLUMES
## Abstract Based on the lower‐bound shakedown theorem by Melan, a method to analyse pavements under cyclic, in particular, rolling contact loading is presented. Repeated sliding/rolling line contact as well as repeated stationary contact is considered. The material is assumed to be rate‐independen
We addres some of the issues that arise when an interior point method is used to handle the master problem in a decomposition approach. The main points concern the efficient exploitation of the special structure of the master problem to reduce the cost of a single interior point iteration. The parti
provide an asymptotic analysis of a primal-dual algorithm for linear programming that uses modified search directions in the final iterations. The algorithm determines the search directions by solving the normal equations using the preconditioned conjugate gradient algorithm. Small dual slack variab