Affordable Access

Publisher Website

Chapter 6 Estimates of Sensitivity Information Using Penalty Functions

Identifiers
DOI: 10.1016/s0076-5392(08)60655-4
Disciplines
  • Computer Science

Abstract

Publisher Summary This chapter discusses the estimates of sensitivity information using penalty functions. It is desirable to exploit the fact that it is possible to estimate sensitivity information from information generated by a solution algorithm as a solution is approached. It seems apparent that any solution algorithm must generate, or at least manipulate, an abundance of information that is relevant to sensitivity and stability analysis. It follows that the class of algorithms based on twice-differentiable penalty functions can readily be adapted to provide estimates of the sensitivity information. The chapter confines analysis to a class of penalty functions; it is reiterated that any valid solution algorithm utilizes and generates analogous information that can be manipulated to provide corresponding sensitivity estimates. For convenience and specificity the results are given in terms of the logarithmic barrier function combined with a quadratic penalty term to handle the equality constraints.

There are no comments yet on this publication. Be the first to share your thoughts.