BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Seminar by Prof. Erick Delage\, HEC Montréal
DTSTART:20150722T120000
DTEND:20150722T133000
DTSTAMP:20260506T084448Z
UID:da656d2b4fc4efac5f9356203ebb9fc73005c224d5e6dc6f9f6b87f5
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Erick Delage\, HEC Montréal\n"Linearized Robust Counter
 parts of Two-Stage Distribution Problems"Abstract\nWe study two-stage dist
 ribution problems wherein some (explicit or implicit) decisions are implem
 ented only once the demand for products has been revealed. Since the robus
 t optimization formulation of this problem is computationally intractable 
 we propose a conservative tractable approximation scheme for this problem 
 based on linearizing the cross terms that appears due to the recourse prob
 lem. We relate this new scheme to methods that are based on exploiting aff
 ine decision rules. Furthermore\, we show that our proposed method can be 
 exploited to provide exact solutions in a family of robust multi-item news
 vendor problem. Using a robust facility location problem\, we also show ho
 w our proposed method can be used to derive conservative approximations th
 at are tighter than existing tractable methods. This is joint work with my
  PhD student Amir Ardestani-Jaafari.Bio\nErick Delage completed is Ph.D. a
 t Stanford University in 2009\, is currently associate professor in the De
 partment of Decision Sciences at HEC Montréal and was recently appointed 
 as chairholder of Canada Research Chair in Decision Making under Uncertain
 ty. He is currently visiting CDM until August 31st 2015.
LOCATION:EPFL\, ODY -1.0019.1 http://plan.epfl.ch/?zoom=19&recenter_y=5863
 773.78544&recenter_x=731540.75234&layerNodes=fonds\,batiments\,labels\,inf
 ormation\,parkings_publics\,arrets_metro\,transports_publics&floor=-1&q=OD
 Y_-1.0019.1
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR