Check out our interactive example notebooks in Google Colab to find out just how simple it is to solve multistage stochastic programming problems with QUASAR®!
What to expect in this notebook? In this example notebook, we are going to look at one of the oldest use cases for stochastic-dynamic programming: reservoir management.
Optimizing the dispatch of hydropower over time is a common activity in power systems planning. In many hydro-thermal systems with a high degree of hydropower, storing too little water may lead to excessive usage of expensive thermal power plants (or even outages) later, whereas storing too much power may lead to spills and thus energy going wasted. Finding the right value of water at any given point in time is therefore a problem of practical importance for many hydropower companies.
While modeling reservoir management as a decision problem is rather standard, this notebook demonstrates how to fit a scenario lattices directly to historical inflow time series, thereby skipping the often tedious step of fitting the right time series model. Fitting lattices directly to data is an innovative feature of QUASAR and particularly well-suited in situations where the random process is stationary and historical data is plentiful.
The notebook also demonstrates on how to obtain water values and how to simulate the optimal decision policy to obtain scenarios of how reservoir content evolves over time.
Are you are interested in seeing a more production-ready model in action? 
Any QUASAR® model can be transformed into a fully-fledged software solution for end users with QUASAR® Cloud. 
