"Most real-world problems involve uncertainty."
The word "stochastic" evokes uncertainty in more ways than one. For most of us, uncertainty arises because we don't know what it means. For optimization experts, like Gyana Parija, stochastic optimization is all about using scenario trees to support decision-making in uncertain environments.
By assigning probabilities to the factors driving outcomes, stochastic programming approaches model how limited resources can meet tomorrow's unknown demands. In this way, the risks and rewards of various tradeoffs can be explored. Gyana has been using stochastic programming methods for years to solve problems such as portfolio allocation, dynamic pricing, inventory management and strategic budgeting.
Stochastic programming offers greater modeling power and flexibility, but it comes at a cost - processing time. With stochastic approaches there is more input data to be managed, the optimization problems are very large and, when there are many variables, the solutions become huge and more difficult to analyze. As a result, stochastic programming has benefited from the development of more efficient algorithms and faster computer processors.
What excites Gyana today is that the advances in processing and algorithms make it possible to begin thinking of replacing traditional forecasts with stochastic models. This means that rather than predicting a limited future using forecasting, decisions can be made that support a wide range of probable scenarios - a more opportunistic approach.
One of the more uncertain factors in many decisions is human behavior. As stochastic models become more sophisticated, Gyana is able to infuse the models he builds with "human" factors, such as politics, custom and culture. By considering human behavior in his models, the results are less uncertain and more accurate and acceptable.
When Gyana isn't busy making problems less uncertain, he loves traveling to interesting places. Luckily, he wasn't uncertain about when to stop chasing the polar bears (to get a picture or two) in Barrow, Alaska or else he wouldn't be here today.
| |
