Scenario Set Partition Dual Bounds for Multistage Stochastic Programming : A Hierarchy of Bounds and a Partition Sampling Approach

@inproceedings{Boland2016ScenarioSP,
  title={Scenario Set Partition Dual Bounds for Multistage Stochastic Programming : A Hierarchy of Bounds and a Partition Sampling Approach},
  author={Natashia Boland and Ilyas Bakir and Brian Dandurand and Alan L. Erera},
  year={2016}
}
We consider multistage stochastic programming problems in which the random parameters have finite support, leading to optimization over a finite scenario set. We propose a hierarchy of bounds based on partitions of the scenario set into subsets of (nearly) equal cardinality. These expected partition (EP) bounds coincide with EGSO bounds provided by Sandıkçı et al. (2013) in the case that the subset cardinality divides evenly into the cardinality of the scenario set, and otherwise interpolate… CONTINUE READING

References

Publications referenced by this paper.
SHOWING 1-10 OF 77 REFERENCES

M-smiplib: A multistage stochastic mixed-integer programming test set library

G. L. Zenarosa, O. A. Prokopyev, A. J. Schaefer
  • http://www.cs.cmu.edu/~gzen/m-smiplib/. Accessed: 2016-10-17. Zenarosa, G. L., Prokopyev, O. A., and Schaefer, A. J. (2014b). Scenario-tree decomposition: bounds for multistage stochastic mixed-integer programs. Optimization Online.
  • 2014
VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

Introduction to Stochastic Programming

VIEW 3 EXCERPTS
HIGHLY INFLUENTIAL