Page 153 - Big Data Analytics for Connected Vehicles and Smart Cities

134 Big Data Analytics for Connected Vehicles and Smart Cities What Are Analytics? 135

items such as energy demand and retail transactions. A full-fledged, multisource
data lake will not be created overnight but will evolve from supporting a few
important primary use cases into a wider capability as other data is added, and
the capability of the data lake is extended. One of the important elements of a
data lake is the ability to draw data together from across an organization and an
enterprise and create an enterprise-wide view of the data. In many transporta-
tion agencies, data is collected in silos and may even be stored for the use of
individual staff members. This type of fragmented stovepipe data storage makes
it very difficult to get the best value for the money from analytics. Fortunately,
however, it is possible to create a virtual data lake, in which data resides in its
original location but is indexed and accessible to the central repository. Chapter
9 details this subject.

6.9 How to Identify Data Needs Associated with Analytics

There is a chicken and egg problem that must be addressed with respect to
the use of analytics. What comes first—the data or the analytic? The answer is
typically that the analytics will be created first of all on the basis of needs that
have been identified. However, it is not possible to conduct an analytic when
the data required is not available. Therefore, in practice the initial list of data
analytics would be filtered to ensure that the early analytics have the required
data available. It is also the case that objectives are linked to use cases, which
are linked to analytics. The best approach would be to pick one or two use
cases that deliver clear and immediate value to the end user and for which any
necessary data is available. The initial pass on the analytics should deliver value
to the user through the use of actionable insights. The results of the initial ana-
lytics should also provide the business justification for further investments of
time and money. It may also be the case during the initial analytics that ways
in which the results can be improved through better data will be revealed. This
sets the scene for the development of a structured data acquisition and use plan.

6.10 Summary

This chapter discusses the nature and characteristics of analytics from a trans-
portation perspective. It presents a formal definition of the term analytic and
interprets it for use in transportation and smart cities. In addition, the chapter
addresses the difference between reporting, analytics, and KPIs, illustrating the
point that all three are intended to work together for a smart city, while ana-
lytics reveal detailed insights into trends and patterns and create the basis for
actionable insights that can influence the performance of smart city transpor-
tation services. Moreover, the chapter discusses the value of analytics in terms