Page 238 - Big Data Analytics for Connected Vehicles and Smart Cities

218 Big Data Analytics for Connected Vehicles and Smart Cities Practical Applications and Concepts for Transportation Data Analytics 219

Scientific traffic engineering would ultimately lead to the design of traffic-
management strategies driven by data and effects. One could also ultimately
foresee a day when design documents could be rewritten to take account of this
new scientific approach. In parallel with the adoption of scientific traffic engi-
neering, it would also be possible to take a scientific approach to the planning
for future investments in transportation facilities including intelligent trans-
portation systems. Access to an integrated data set covering multiple aspects of
transportation service delivery and prevailing conditions would also support
before studies to be conducted at any point in the span of the data set. Identify-
ing and understanding the exact effects of investments made in the past would
lead to finely targeted investments in the future. This could well be the genesis
of a new scientific approach to transportation planning and engineering.
A thought related to the scientific approach lies in the need to define,
from a data perspective, terms that are typically used by transportation people
on a subjective basis. These include terms like recurring and nonrecurring con-
gestion and bottleneck, as illustrated in this chapter.
A final thought related to the freeway speed variability work concerns
the availability of performance data beyond the boundaries of the agency. The
availability of such data sources could put pressure on local agencies regarding
operational efficiency. As an example, during the freeway work the data science
team noted the occurrence of an incident on September 20, 2014 between 3
and 4 p.m. In discussion with the client team, it turned out that this was the
time and date of a major snowstorm. The significance in this interaction lies
in the fact that an independent data science team could identify performance
issues with no recourse to public agency data.

Accessibility Index
An interesting aspect of the accessibility analytic lies in the ability to use ob-
served data where typically synthetic data from a transportation model would
be used. There is a possibility that the observed data could be combined with
this synthetic data to provide a more complete picture of the smart city. How-
ever, it would be necessary to consider the accuracy of the resultant data set,
since data from the transportation model is likely to be based on a much smaller
sample size than the observed data.
Toll Return Index

Once the toll return index framework has been established, it is possible to use
in-depth analytics to understand how changes in safety, efficiency, and user
experience will affect the toll return index. While the toll return index has been
calculated for an entire toll network in this example, it would also be possible
to explore the dynamics of the toll return index for different segments of the
total network to understand the variations in safety, efficiency and user experi-