EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
385. Scaled Cuts for Multistage Stochastic Mixed-Integer Programs
Invited abstract in session TA-35: Optimization under uncertainty: theory and solution algorithms, stream Stochastic, Robust and Distributionally Robust Optimization.
Tuesday, 8:30-10:00Room: 44 (building: 303A)
Authors (first author is the speaker)
| 1. | Ward Romeijnders
|
| Department of Operations, University of Groningen |
Abstract
We consider Benders' decomposition algorithms with affine optimality cuts for multistage stochastic mixed-integer programs. Exploiting extended spaces, we derive a hierarchy of convex polyhedral lower bounds for the expected cost to-go functions in the problem. These lower bounds, however, are not tight for mixed-integer state variables in general. That is why, we also introduce so-called scaled cuts. The advantage of these scaled cuts is that they allow for parametric non-linear feasibility cuts in later time stages, but that the optimality cuts for the expected cost to-go functions remain linear. We establish convergence by proving that the optimality cuts in the first stage recover the convex envelope of the first-stage expected cost to-go function.
Keywords
- Programming, Stochastic
Status: accepted
Back to the list of papers