Page 234 - Big Data Analytics for Connected Vehicles and Smart Cities

214 Big Data Analytics for Connected Vehicles and Smart Cities Practical Applications and Concepts for Transportation Data Analytics 215

education, healthcare, and retail opportunities. Movement analytics data can
also be blended with U.S. census data to provide additional insight into the
characteristics of each zone. While this approach still contains an element of
estimation, it is based on observed data that can consider the dynamics of ac-
cessibility over time.

Analytics Used
The primary analytic used is known as the accessibility index and it is com-
prised of the following components:

• Estimated travel time between residential and opportunity zones: Zones
within the smart city region are defined, based on U.S. census tract
boundaries as census data will be required to define demographics for
each zone. Movement analytics from smart phone applications will be
used to characterize travel between zones and to identify overnight resi-
dential locations for smart phone users. Opportunity zones will be iden-
tified based on U.S. census data. Travel time analytics will be determined
to illustrate average travel times between residential zones and opportu-
nity zones. This analytic will be a factor in the overall accessibility index.
• Estimated travel time reliability between residential and opportunity zones:
By considering travel times between zones over a suitable period it is
possible to develop an estimate of travel time reliability between residen-
tial and opportunity zones. This analytic will also be a factor within the
overall accessibility index.
• Cost of travel between residential and opportunity zones as a proportion of
household income: U.S. census data would be used to estimate household
income for each residential zone. Cost of travel between residential and
opportunity zones will be determined based on the travel time and travel
time reliability between zones and an assumed hourly cost of travel. This
latter figure is typically related to the average earnings per hour for the
city, but it could also consider the average earnings per hour for each
zone. The three analytics are combined into a single accessibility index,
which is defined as: accessibility index from zone 1 to zone 2 = travel
time between zones, travel time variability between zones, and the cost
of travel as a proportion of household income between zones. Note that
it would also be possible to apply the type of generalized cost modeling
used in transportation simulation models as an alternative to this ap-
proach.