BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:event-916
DTSTAMP:20260408T013020Z
SUMMARY:SMT Lunch Lecture Breno Alves Beirigo
DESCRIPTION:A business class for autonomous mobility-on-demand: Optimizatio
 n- and learning-based strategies to model service quality contracts in rid
 esharing systems With the popularization of transportation network compani
 es (TNCs) (e.g.\, Uber\, Lyft) and the rise of autonomous vehicles (AVs)\,
  even major car manufacturers are increasingly considering themselves as a
 utonomous mobility-on-demand (AMoD) providers rather than individual vehic
 le sellers. However\, matching the convenience of owning a vehicle require
 s providing consistent service quality\, taking into account individual ex
 pectations. Typically\, different classes of users have different service 
 quality expectations\, especially in terms of reliability and responsivene
 ss. Nonetheless\, planning models presented in the AMoD literature do not 
 enable active control of service quality\, sometimes allowing extensive de
 lays and user rejections. In this study\, we propose both learning- and op
 timization-based strategies to actively control service quality in AMoD sy
 stems\, increasing and decreasing the number of used vehicles in the short
  term to meet diversified user expectations. We have used these expectatio
 ns to establish service quality contracts\, allowing heterogeneous users t
 o choose ride experiences that best match their preferences. Based on an e
 xperimental study using New York City taxi data\, we show how providers ca
 n adequately cater to each segment of the customer base without necessaril
 y owning large fleets through a service-quality-oriented on-demand hiring 
 approach.   Zoomlink will be posted
DTSTART:20210609T103000Z
DTEND:20210609T113000Z
LOCATION:Zoom
END:VEVENT
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
DTSTART:20210609T103000
TZNAME:UTC
TZOFFSETTO:+0000
TZOFFSETFROM:+0000
END:STANDARD
END:VTIMEZONE
END:VCALENDAR