Abstract
We propose a class of partially observable multistage stochastic programs and describe an algorithm for solving this class of problems. We provide a Bayesian update of a belief-state vector, extend the stochastic programming formulation to incorporate the belief state, and characterize saddle-function properties of the corresponding cost-to-go function. Our algorithm is a derivative of the stochastic dual dynamic programming method.
Original language | English (US) |
---|---|
Pages (from-to) | 505-512 |
Number of pages | 8 |
Journal | Operations Research Letters |
Volume | 48 |
Issue number | 4 |
DOIs | |
State | Published - Jul 2020 |
Funding
This article was developed, in part, based upon funding from the Alliance for Sustainable Energy, LLC, Managing and Operating Contractor, USA for the National Renewable Energy Laboratory for the U.S. Department of Energy. The authors thank three referees whose comments improved the paper.
Keywords
- Bayesian
- Multistage
- Partially observable
- Stochastic dual dynamic programming
- Stochastic programming
ASJC Scopus subject areas
- Software
- Management Science and Operations Research
- Industrial and Manufacturing Engineering
- Applied Mathematics