Eric Horvitz Microsoft Research CCC RISES Washington DC, Feb. 2011 - - PowerPoint PPT Presentation

eric horvitz
SMART_READER_LITE
LIVE PREVIEW

Eric Horvitz Microsoft Research CCC RISES Washington DC, Feb. 2011 - - PowerPoint PPT Presentation

Eric Horvitz Microsoft Research CCC RISES Washington DC, Feb. 2011 Transportation Work distribution Green computing Datacenter efficiencies Energy usage forecasting, tracking, controls Tools for others ~175,000 people ~55,000 in Puget


slide-1
SLIDE 1

Eric Horvitz Microsoft Research

CCC RISES Washington DC, Feb. 2011

slide-2
SLIDE 2

Transportation Work distribution Green computing Datacenter efficiencies Energy usage forecasting, tracking, controls Tools for others

slide-3
SLIDE 3
slide-4
SLIDE 4

~175,000 people ~55,000 in Puget Sound Region

slide-5
SLIDE 5

Copenhagen meeting Emissions abatement  450 ppm by 2030 (~2

  • ).

(IEA 10/09)

slide-6
SLIDE 6

Matching algorithms: planned and instant Incentives: mechanism design for truthful reporting Collaboration & plans with related goals Preferences and comfort: social component Daily workflow: Outlook/Exchange

Coordination with King County Metro, WashDOT, MS Facilities, MS Sustainability.

More details: Collaboration and Shared Plans in the Open World: Studies of Ridesharing, IJCAI 2009.

slide-7
SLIDE 7

GPS data: AM/PM commutes to & from Microsoft

With E. Kamar

slide-8
SLIDE 8

Cost-benefit

Earlier departure Delayed arrival Increased travel Savings on effort Fuel, environment

Start time to +

  • Start time

to +

  • Arrive time

to +

  • Arrive time

to +

  • D Trip Duration

to +

  • D Trip Duration

to +

  • Shared & divergent preferences
slide-9
SLIDE 9

Arrive time to +

  • Arrive time

to +

  • Cost-benefit

Earlier departure Delayed arrival Increased travel Savings on effort Fuel, environment

Start time to +

  • Start time

to +

  • D Trip Duration

to +

  • D Trip Duration

to +

  • Shared & divergent preferences
slide-10
SLIDE 10
  • Assignments based
  • n observed trips.
  • Cost-benefit
  • Departure change
  • Delayed arrival
  • Increased travel
  • Savings on effort,

fuel, environment

slide-11
SLIDE 11
slide-12
SLIDE 12

Rideshare queued Commute request Rideshare starts Single rider starts

slide-13
SLIDE 13

Rideshare queued Commute request Rideshare starts Single rider starts

slide-14
SLIDE 14

Normal commute Computed rideshares

slide-15
SLIDE 15

10 20 30 40 50 60 108 215 430 860

System Efficiency Number of agents

Efficiency on number of commutes Efficiency on total cost

Number of participants 

"What If?" Studies

slide-16
SLIDE 16

Fuel Cost 

"What If?" Studies

slide-17
SLIDE 17

Best Park & Ride Locations?

slide-18
SLIDE 18

Challenge: Understanding acceptance, perceptions, social considerations

Address concerns, leverage opportunities Trusted organizations Referral, reputation

  • e.g., existing online social networks (e.g., link distance bounds)

?

slide-19
SLIDE 19
slide-20
SLIDE 20

Autonomous vehicles? …Yes. But…preferences, incentives, optimization!

  • Direction: Public microtransit