Page 220 - Big Data Analytics for Connected Vehicles and Smart Cities

200 Big Data Analytics for Connected Vehicles and Smart Cities Practical Applications and Concepts for Transportation Data Analytics 201

key objectives of the work was to demonstrate the use of analytics on a large
observed data set, it was decided to adopt the latter approach.

Initial Analytics Approach—Speed Variability

The analysis makes use of privately sourced Inrix speed data collected from
2013 to 2014 on the subject freeway, approximately 36 miles in each direc-
tion. There are 138 TMC locations for which Inrix [1] data was available on a
minute-by-minute basis.
While direction signs on the freeway make use of formal directional des-
ignations based on compass headings at various locations of the freeway (east/
west or north/south), for purposes of the analysis, the data was arrayed in clock-
wise and counterclockwise directions, proceeding clockwise from mile marker
9.97 in the west to mile marker 46.3 in the east, and vice versa proceeding
counterclockwise from mile marker 46.5 in the east to mile marker 10.0 in the
west. Note that the data analytics were conducted in Teradata Aster, with sum-
maries and visualizations developed in Tableau.
Figure 10.2 shows a summary of the available data for the two directions
for the full year and illustrates the speeds when considering only weekdays and
then only weekday peak periods. An analysis of the data was conducted to

Figure 10.2 Summary of available data for two directions for the full year.