Page 172 - Big Data Analytics for Connected Vehicles and Smart Cities

152 Big Data Analytics for Connected Vehicles and Smart Cities The Practical Application of Analytics to Transportation 153

experience from the early analytics providing input and recommendations for
the improvement of data acquisition and data collection for subsequent analy-
sis. This chapter intends to build on the information in Chapter 6 by providing
some practical insight into how analytics can be defined and applied for trans-
portation service delivery within a smart city and initiative or context.
It should also be noted that some analytics will be discovered during early
analytics work. The nature of the analytics process allows for a certain amount
of discovery, and the latest approach to data and information management is to
find ways for the data to speak. A skilled analyst working with a combination
of data sets within a data lake is highly likely to uncover some new relationships
that will lead to the definition of new analytics. Therefore, it is probable that a
range of predefined analytics will be used to support the initial analytics work
and that these will be supplemented with additional analytics that are discov-
ered during the analytics work. This provides the potential for breakthroughs in
understanding and insight as new analytics, new connections, and new mecha-
nisms are identified and defined because of the combined data set or data lake.
This feature has the scope for considerable innovation on the interface between
transportation and data science and is likely to be a fertile subject area for re-
search and the practical application of analytics to transportation.
The application of analytics is likely to take place within a wider context
of evaluation and understanding the effects of transportation investments. It is
useful to consider analytics, along with the data lake, as tools that work very
well together, with the availability of data enabling a rich set of analytics.
As a final note in this chapter, it is also worth considering the future role
of analytics within a path toward total automation of the back office. In fact,
the ease with which data can be sourced for the purposes of smart city transpor-
tation analytics is significantly impacted by the formation of a data lake. Cur-
rently, data is often kept in a fragmented, poorly cataloged form. Merging data
and having an organization-wide view of available data is an important step in
enabling the analytics discussed in this chapter.
There is considerable effort and interest in the concept of autonomous
vehicles, and this could be considered as the application of automation to one
component of the overall vehicle and the highway infrastructure. It is reason-
able to assume that the back-office component of transportation systems will
also be subject to the same level of automation. Today, we have transportation
management center operators and managers reviewing and evaluating data re-
garding current transportation conditions, in some cases with the help of so-
phisticated decision-support systems.
Future traffic management systems could feature a higher level of automa-
tion, based on a more detailed understanding of causes and effects. This defines
the role for analytics in the future transportation system. We can start now by
identifying analytics that characterize transportation conditions both now and