BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Challenging Computer Science Problems at Ocado - Dr Vince Darley\,
  Head of Analytics &amp\; Optimisation\, Ocado
DTSTART:20110126T141500Z
DTEND:20110126T151500Z
UID:TALK28687@talks.cam.ac.uk
CONTACT:Stephen Clark
DESCRIPTION:While Ocado has many challenging problems in hardware\, roboti
 cs\, distributed systems\, iPhone/Android\, web technologies and so on\, i
 n this talk I’ll describe and formulate a handful of the optimisation an
 d modelling problems we deal with\, and their similarities and differences
  to the archetypal computer science problems. These range across:\n\n•	o
 ptimising our delivery routes (a kind of TSP or VRP)\n\n•	packing our cu
 stomers orders (a kind of 3d bin-packing)\n\n•	calculating optimal drivi
 ng paths (Dijkstra shortest path)\n\n•	forecasting product demand (machi
 ne learning)\n\n•	simulating/optimising our warehouse (a discrete event 
 simulation with many different embedded optimisation sub-systems)\n\n•	o
 ptimising our website product recommendations\n•	helping customers shop 
 more quickly by suggesting predictions of what they will buy\n\nOcado’s 
 scale (1 million items picked per day\, 100000 \norders delivered per week
 ) ensures that in all of these areas\, optimising to attain the last fract
 ion of a percent of improvement is of genuine value\, and not just a theor
 etical nicety. However all of these problems have additional complications
  and constraints which must somehow be included in the theoretical formula
 tion.\n\nBIO\n\nDr Darley started his career with a Maths degree from Trin
 ity College\, Cambridge\, followed by a move to Harvard for a Masters in C
 omputer Science and a PhD (jointly between Applied Maths\, Computer Scienc
 e and Economics)\, including a spell at the Santa Fe Institute. His resear
 ch studied both the dynamics of groups of “economic agents” when tradi
 tional assumptions of perfect rationality\, complete information\, etc. ar
 e relaxed\, and phase transitions in problem difficulty of NP-complete pro
 blems.  He then spent a few years working at Bios Group in Santa Fe\, mode
 lling and optimising for organisations such as Unilever\, P&G\, Ford\, Nas
 daq. He wrote a book summarising several years of research on the Nasdaq s
 tock market\, including accurate predictions of changes to market dynamics
  which would come to be caused by “decimalisation”. In 2000 he moved b
 ack to the UK to set up his own software and consulting business which pro
 vided sophisticated modelling and optimisation to European businesses. Sin
 ce early 2009 he has been chief scientist with Ocado.\n
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR