Page 147 - Building Big Data Applications

P. 147

Travel and tourism industry

applications and usage

Travel and big data

The global travel industry is expected to grow to 10% of Global GDP by 2022, or an
annual revenue of around $ 10 trillion. This industry is at a tipping point where the
implementation of prescriptive and predictive analytics, artiﬁcial intelligence and neural
networks can improve the actions, increase revenue, get more traction with the
customers, and overall transform the industry to its next iteration of growth.
In a long running tradition in the industry, travel companies are known for capturing
and storing massive amounts of data. During every step of every travel journey they
collect data including customer data, ﬂight paths, driving paths, bus journeys, train
journeys, transactions, driving styles, check-ins etc. CRM packages implemented at
hotels collect vital data from customers for both regulatory compliance and the
marketing requirements, and let us not forget yield revenue management was invented
in the travel industry already years ago. The data collected sounds very interesting but
the business value from this data was difﬁcult to deliver and analyze due to the limi-
tations of the underlying infrastructure. The evolution of computing platforms driven by
eCommerce companies and the sheer amount of processing power, cheap, and powerful
storage solutions such as Hadoop and many technology vendors waiting to help out, this
information can ﬁnally be put to use to make the customer feel more appreciated and
better serviced, resulting in more revenue and higher proﬁts.
The big question that you want to ask but are unsure of the answers is how do I get
value delivered from this data? The answer to this question is to ﬁrst layout the different
blocks of data that you need to use for answering a question whether business value or
not. The data blocks will provide you an insight on the interconnectedness of the data;
the missing pieces will reveal themselves and provide you opportunity to create
a connected data chain. Once this exercise is complete, then you set out to build the
application required to answer the question in your mind, in fact my suggestion is to
start with a search bar application to see how much of intuitive reply you will get for
a search and then start exploring the system. Fig. 8.1 shows the data block layout.
The data block can be built as pyramid layers with increasing order of summarization
and aggregation. This layout is essential if you are looking at big data which can be
chaotic if not understood well.

142 143 144 145 146 147 148 149 150 151 152