Robust Optimization with Decision Dependent Uncertainty

In this project, we investigate optimization problems where the parameters are uncertain and the uncertainty can be altered by the decisions. In most of the practical applications where optimization problems arise, the parameters are uncertain and there is not sufficient historical data to come up with reliable estimates. Robust optimization is used to hedge against this uncertainty and to make decisions that remain feasible for all possible realizations of parameters and that perform best in the worst case. Most of the studies in robust optimization take the uncertainty to be independent of the decisions. However, it is often possible to alter/reduce uncertainty at an additional cost. Robust optimization with decision dependent uncertainty is a recent field of research with many possible real life applications. The resulting problems are significantly more difficult than their deterministic counterparts. The aim of this project is to investigate the applications and difficulty of these problems and to propose effective solution methods.

Keywords:optimization

Disciplines:Applied economics, Economic history, Macroeconomics and monetary economics, Microeconomics, Tourism, Applied mathematics in specific fields, Statistics and numerical methods, Other philosophy, ethics and religious studies not elsewhere classified, Theory and methodology of philosophy, Philosophy, Ethics, Business administration and accounting, Management