Methods for robustness programming

N.J. Olieman

Research output: Thesisinternal PhD, WU

Abstract

Robustness of an object is defined as the probability that an object will have properties as required. Robustness Programming (RP) is a mathematical approach for Robustness estimation and Robustness optimisation. An example in the context of designing a food product, is finding the best composition of ingredients such that the product is optimally safe and is satisfying all specifications. Another example is the investment in a portfolio of stock market shares. The number of shares to invest in is typically a controllable factor. The future shares prices and resulting portfolio return are typically uncontrollable factors. It is interesting to find the composition of shares for which the probability of reaching a predefined target return is as high as possible.
In this research alternative methods for Robustness Programming are developed with favourable optimisation properties for finding a design with a Robustness as high as possible. Some of these methods are generally applicable, while other methods use specific problem characteristics. A framework for Robustness Programming is developed for modelling design problems from a wide application area and to select the applicable RP methods for such design problems.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
Supervisors/Advisors
  • van Beek, Paul, Promotor
  • Hendrix, Eligius, Co-promotor
Award date12 Mar 2008
Place of PublicationS.l.
Print ISBNs9789085048763
Publication statusPublished - 2008

Keywords

  • mathematics
  • operations research
  • estimation
  • programming
  • monte carlo method
  • computational mathematics

Fingerprint Dive into the research topics of 'Methods for robustness programming'. Together they form a unique fingerprint.

  • Cite this