Historically, progress on these techniques has been slow, because (a) exact approaches lead to nonlinear equations for which numerical solution methods typically execute slowly and may fail to converge to a desirable shape, while (b) linear approximations, though fast, are often complicated to derive and may also yield undesirable shapes. Nevertheless, recent research is producing interesting, albeit limited, results. In this presentation, the speaker will give a status report on such methods, focusing mainly on variational techniques.
Ramon F. Sarraga
Analytic Process Department
General Motors Research and Development Center