The importance of incorporating uncertainty into optimization problems has always been known; however, both the theory and software were not up to the challenge to provide meaningful models that could be solved within a reasonable run time. Over the last 15 years, the continuous improvements made to the theoretical as well as the algorithmic area of stochastic and mixed integer linear optimization have changed this situation dramatically. In this recording, we focus on stochastic optimization models and easy-to-understand algorithms, amenable to being easily solved with Gurobi. The intended audience for this webinar includes those with a background in optimization and knowledge on basic probability and statistics. This 35-minute video recording consists of:
- A quick introduction to stochastic optimization
- Types of stochastic optimization problems
- Types of models that can be solved easily: two-stage stochastic problems with expected value and coherent risk measures
- Overview of the main algorithms: sample average approximation
- Examples of common problems with Gurobi
You can download the materials associated with this webinar here.