Providing Engineering Services With Smart Objects: An Active Big Data Approach

Introduction

The expansion of the Big Data concept has affected the changing and evolving nature of data. Big Data has matured as a critical component of the Information Systems, Computer Science, Information Processing, and Data Analytics disciplines. Research has focused for a number of years on characterizing Big Data with five Vs: volume, velocity, variety, veracity, and value as originally described by Kaisler, Armour, Espinosa, and Money (2013). These characteristics are recognized and discussed in the information systems research literature (Goes, 2014) as being valuable to the information systems field. The Vs aid researchers and practitioners in framing approaches for improving systems by focusing research on how to addressing the challenges and difficulties associated with Big Data. They improve our understanding of Big Data and clarify the issues, methodological concerns, and complex research problems facing practitioners and users of Big Data.

The descriptive characterization of Big Data offered in 2013 was the authors’ proposed approach to the problem of communicating about the state of Big Data’s five Vs. The paper advances our understanding of the different ways researchers and practitioners approach and operationalize the Vs used to describe Big Data. Further, it clarifies how the VS could be addressed from a research perspective (at that time, 2012-13). The operationalization components of the Vs evolved from the authors’ appreciation and understanding of how the simple idea of Big Data morphed to mean different things over time. It first referred only to a volume of data that could not be efficiently processed using the available system methods, tools, and technologies. But this original pragmatic definition offered little insight into the solutions to the problems associated with collecting, handling and processing Big Data. It was incapable of dealing with data growth because the growth never stopped, or even slowed. As new expanded storage methods and processing performance improved, data accessibility exploded (again and again) because if data could be more easily accessed, more could and would be collected. Researchers could simply not keep up with growing data volumes, but researchers and users wanted to communicate about what they were doing about the “growth of data” and share solution attempts, successes, and failures. A second issue (beyond ever-expanding volumes) also became important because data definitions which first focused on structured data needed to address massive amounts of unstructured text and imagery. The explosion of these data was not accompanied by a corresponding new storage medium. Almost 20 years ago, Laney (2001) described the Big Data occupying this environment with 3 Vs (volume, variety, and velocity) and helped initiate the direction of research in the field toward an analysis of attributes and characteristics that if better managed could deliver greater value to the users of Big Data. Big Data, throughout this period, is viewed as being inert. The three, and subsequent five Vs thus describe Big Data using passive attributes that are only derived or changed through manipulation by external events and procedures.

Big Data (without a widely accepted definition, not well understood, and not clearly characterized) was being noticed in many areas because it was associated with potential business and process value. For example, Big Data was recognized as a national challenge and priority along with healthcare and national security in 2010 by the American Institute of Physics (2010). Two years later the National Science Foundation led a large multi-agency effort and research program to develop new Big Data tools and techniques, and advance the technological understanding of the massive amounts of information and knowledge available from Big Data. The goals of this effort were to apply the Big Data findings and tools to health, energy, defense, education and researcher goals (Mervis, 2012).

This paper now posits that the proposed characteristics of “inert” Big Data as viewed in 2013 must again evolve and possibly incorporate new Big Data characteristics. The concept of Smart Data, as proposed in this paper, imbues Big Data with additional attributes (beyond the 5 Vs) and characteristics through the application of analytics.