Big Data Market 2015 - From Prediction To Optimization - HPC ASIA
Features

Big Data Market 2015 – From Prediction To Optimization

4096609612- (1)
Kirk Borne

Kirk Borne

Surveys of businesses reveal their goals, use cases, and priorities for big data. Surprisingly, one poll showed in approximately 10% of businesses surveyed that “big data is not of interest” to their organization. These businesses are apparently satisfied with inefficiencies, non-optimal revenue streams, and falling behind their competition. We assume that your business is not one of those. So, here are a few of the survey findings and their relationship to the common analytics use cases in various business sectors that are driving the big data market this year.

Nearly 60% of all business place at least one of these three big data outcomes within their top 3 expectations: (a) impact on customer relationships; (b) redefine product development; or (c) change their operational organization.  Half of all businesses place at least one of these big data outcomes within their top 3 expectations: (1) make the business more data-focused; or (2) optimize their supply chain.

The top 3 business big data use cases are customer analytics (48%), operational analytics (21%), and fraud and compliance analytics (12%).  For Asia-Pacific manufacturers, 37% plan to use big data and analytics to improve production quality management, which a focus on cost reduction, improved productivity, and acquisition of new customers. As other studies have found, it is now abundantly clear that big data and analytics are fundamental to next-gen manufacturing.

The business surveys have also discovered the factors that are driving this enormous interest in big data analytics across a large variety of industries and organizations. These factors include: making discoveries across multiple disparate data sources; predicting customer behavior; predicting sales or service requests; predicting fraud or financial risk; analyzing social media for customer sentiment; IT security risk assessment; and analyzing machine logs (from sensors, web traffic, or machines).

It is clear from these results that predictive analytics is a major feature of the big data era that attracts business interest, that motivates business innovation, and that delivers big return on investment.  In all of the top 3 business big data use cases, prediction is key: in customer relationships (e.g., marketing, acquisition, conversion, retention), in operational systems (e.g., product demand forecasting, supply chain management, product placement, machine predictive maintenance), and in risk management (e.g., fraud, compliance, IT security).

Predictive analytics is one of the defining characteristics of advanced analytics, thereby distinguishing it from traditional business reporting (= descriptive analytics, or business intelligence). Businesses want this predictive capability, to improve discovery, decisions, and dividends from their big data collections.  A few more advanced organizations are now moving beyond predictive to the next stage: prescriptive analytics, which is a form of analytics that identifies the optimal course of action that yields the best outcome.

It is no longer sufficient for businesses to explore their data simply to report what happened last season (descriptive analytics), and it will soon be insufficient to use the data to predict what will happen next season (predictive analytics).  For businesses to remain competitive, to address their top use cases and application areas, and to maximize return on investment, it will become imperative for businesses to use their data to discover the best course of action that will deliver optimized outcomes in the customer relationships, operational systems, and risk management processes. Optimization will become the killer app for big data analytics in all phases and functions of the modern digital business.

Author Bio:

Dr. Kirk Borne is a member of the NextGen Analytics and Data Science initiative within the Booz Allen Hamilton Strategic Innovation Group. He is also an advisor for several firms. Previously, he was Professor at George Mason University, where he did research, taught, and advised students in the graduate and undergraduate Informatics and Data Science programs. Prior to that, he spent nearly 20 years supporting large scientific data systems at NASA. Follow him on Twitter at @KirkDBorne.

Comments

comments

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

To Top
error: Content is protected !!
/body /html