|
|
Risk-Averse Optimization in Engineering Design and Management
Decisions often have to be made without entirely knowing their consequences. Typical in these situations is the presence of various random variables, calibrating costs or hazards, that depend on a vector of decision variables which may be chosen with the idea of minimizing those costs or hazards, or at least keeping them at an acceptable level. But the decision can only shape the outcomes of the random variables, rather than produce a definite result. To cope with this, it is common in engineering to minimize expected value, or to put constraints on the probability that a given threshold might be exceeded. But risk-averse modeling has advanced dramatically and nowadays offers approaches with better safe-guarding and at the same time superior capabilities in computation.
R. Tyrrell (Terry) Rockafellar received a PhD in mathematics from Harvard University in 1963 and joined the Mathematics Department of the University of Washington in Seattle in 1966. He officially retired there in 2002, becoming Professor Emeritus, but in 2003 was appointed Adjunct Research Professor in the Department of Industrial and Systems Engineering at the University of Florida.
His research has focused on mathematical challenges in problems of optimization, and has led to over 200 publications and several books.
|
|
|
|