Many engineering systems are affected by uncertainty in future demands or inputs. Decisions regarding their design, however, typically must be made in the present. Two-stage stochastic programming can consider this type of problem but, in the past, procedures to fully incorporate the uncertainty have come with a high computational cost. New algorithmic developments, such as Regularized Stochastic Decomposition (RSD), now allow more complex systems to be considered. This paper provides an overview of the RSD method and its extensions and demonstrates the application of two-stage stochastic programming to two water resources problems with different problem structures and types of uncertainty.
|Original language||English (US)|
|Number of pages||8|
|State||Published - Dec 1 1996|
ASJC Scopus subject areas
- Environmental Science(all)
- Earth and Planetary Sciences(all)