Invited Session Thu.2.MA 141

Thursday, 13:15 - 14:45 h, Room: MA 141

Cluster 22: Stochastic optimization [...]

Two-stage stochastic programming and beyond


Chair: Rüdiger Schultz



Thursday, 13:15 - 13:40 h, Room: MA 141, Talk 1

Dimitri Drapkin
Decomposition methods for optimization problems with stochastic order constraints induced by linear recourse

Coauthor: Rüdiger Schultz


We develop linear programming equivalents for two-stage stochastic optimization models with linear recourse and dominance constraints of first and second order. In the favourable case, where only continuous variables are present in the second stage, cutting-plane decomposition
algorithms are proposed and discussed along with the computational



Thursday, 13:45 - 14:10 h, Room: MA 141, Talk 2

Charlotte Henkel
Some remarks on linear stochastic bilevel programs

Coauthor: Rüdiger Schultz


Compared to linear stochastic two-stage programs, linear stochastic bilevel problems (LSBP) exhibit a strongly increased complexity. Starting from a deterministic linear bilevel problem, we derive structural properties for LSBPs using state-of-the-art parametric optimization techniques. As an outcome, we obtain rather weak analytical results. This significantly effects risk measures and solution algorithms for this kind of problem.
We emphasize our results by instructive examples.



Thursday, 14:15 - 14:40 h, Room: MA 141, Talk 3

Nadine Wollenberg
Stochastic vehicle routing in forwarding agencies

Coauthors: Uwe Clausen, Rüdiger Schultz, Sascha Wohlgemuth


The performance of forwarding agencies handling less-than-truckload freight is mainly influenced by uncertainty in terms of customer demand and travel times. In the talk we discuss two-stage stochastic integer programs with different objective functions such as minimizing the total travel time or minimizing the number of vehicles used for a feasible routing. For the ranking of the resulting stochastic cost profiles we employ different stochastic quality measures leading to risk neutral models and those quantifying some aversion against risk. Algorithmically we rely on scenario decomposition achieved by Lagrangean relaxation of nonanticipativity. Some first computational experiments with realistic problem instances relevant for forwarding agencies in the Ruhr Area are presented.


  The majority of financial loan companies provide the service of getting North Carolina Loans Online for U.S. citizens. Of course, the choice is not that easy, as there exist great number of different preparations. Notwithstanding, Cialis is the one that definitely differs from all other products.