BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Optimization and Incentives Seminar
SUMMARY:Control of Stochastic Processing Networks - R. J.
Williams (Department of Mathematics\, University o
f California\, San Diego)
DTSTART;TZID=Europe/London:20110711T143000
DTEND;TZID=Europe/London:20110711T153000
UID:TALK31476AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/31476
DESCRIPTION:Stochastic processing networks (SPNs) are a signif
icant generalization of conventional queueing netw
orks that allow for flexible scheduling through dy
namic sequencing and alternate routing. SPNs arise
naturally in a variety of applications in operati
ons management and their control and analysis pres
ent challenging mathematical problems. One approac
h to these problems\, via approximate diffusion co
ntrol problems\, has been outlined by J. M. Harris
on. Various aspects of this approach have been dev
eloped mathematically\, including a reduction in d
imension of the diffusion control problem. However
\, other aspects have been less explored\, especia
lly\, solution of the diffusion control problem\,
derivation of policies by interpretating such solu
tions\, and limit theorems that establish optimal
ity of such policies in a suitable asymptotic sens
e.\n\nIn this talk\, for a concrete class of netwo
rks called parallel server systems which arise in
service network and computer science applications\
, we explore previously undeveloped aspects of Har
rison's scheme and illustrate the use of the appro
ach in obtaining simple control policies that are
nearly optimal. Identification of a graphical stru
cture for the network\, an invariance principle an
d properties of local times of reflecting Brownian
motion\, will feature in our analysis. The talk w
ill conclude with a summary of the current status
and description of open problems associated with t
he further development of control of stochastic pr
ocessing networks.\n\nThis talk will draw on aspec
ts of joint work with M. Bramson\, M. Reiman\, W.
Kang and V. Pesic.
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberfo
rce Road\, Cambridge
CONTACT:Elena Yudovina
END:VEVENT
END:VCALENDAR