Abstract:  We discuss recent advances in modeling and computation for time-dynamic optimization under uncertainty via multi-stage stochastic programming. We discuss three modeling ideas, and associated computational tools, that move multi-stage stochastic programming closer to a class of continuous-action continuous-state Markov decision processes. We also point to gaps. This is joint work with Oscar Dowson and Bernardo Pagnoncelli.
 
Bio sketch: David Morton is the Walter P. Murphy Professor in IEMS at Northwestern University. Previously, he was on the faculty in ORIE at UT-Austin, worked as a Fulbright Scholar at Charles University in Prague, and was a postdoc in the OR Department at the Naval Postgraduate School. His research interests include stochastic and large-scale optimization with applications in public health, energy systems, and security.