Decomposition and duality based approaches to stochastic integer programming

Christiansen, J 2018, Decomposition and duality based approaches to stochastic integer programming, Doctor of Philosophy (PhD), Science, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Christiansen.pdf Thesis application/pdf 1.45MB
Title Decomposition and duality based approaches to stochastic integer programming
Author(s) Christiansen, J
Year 2018
Abstract Stochastic Integer Programming is a variant of Linear Programming which incorporates integer and stochastic properties (i.e. some variables are discrete, and some properties of the problem are randomly determined after the first-stage decision). A Stochastic Integer Program may be rewritten as an equivalent Integer Program with a characteristic structure, but is often too large to effectively solve directly. In this thesis we develop new algorithms which exploit convex duality and scenario-wise decomposition of the equivalent Integer Program to find better dual bounds and faster optimal solutions. A major attraction of this approach is that these algorithms will be amenable to parallel computation.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Science
Subjects Optimisation
Keyword(s) Optimisation
Lagrangian duality
Stochastic programming
Mixed-integer programming
Stochastic mixed-integer programming
Lagrange multipliers
Versions
Version Filter Type
Access Statistics: 42 Abstract Views, 44 File Downloads  -  Detailed Statistics
Created: Wed, 27 Mar 2019, 13:39:27 EST by Adam Rivett
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us