Page 151 - Intelligent Digital Oil And Gas Fields

114 Intelligent Digital Oil and Gas Fields

• Variability: this Big Data attribute can occur in each of the other six pil-
lars. The range of variability (not to be confused with the more widely
adopted uncertainty) depends on the stage of an E&P program under
consideration as a function of time, location, or some other measured
parameter. Moreover, Begg et al. (2014) refer uncertainty to not know-
ing the value (or answer) of some quantity and define variability as the
multiple values the quantity has at different locations, times or instances.
• Value: this attribute clearly represents the most important component of
Big Data, measured in monetary or nonmonetary capacity.
In the transcript of a recent survey by Accenture and Microsoft, focusing on
2016 E&P digital trends, Holsman and Richards (2016) have reported that
almost 90% of responding E&P companies, despite the industry downturn,
plan to maintain or increase investments in digital technologies, predomi-
nantly in Big Data-powered analytics, the IoT, and cloud-enabled mobility.
However, though the survey results indicate that more than half of respon-
dent believe that digital technologies have added significant value to their
businesses, the general impression is that Big Data analytics are still being
widely underutilized in the oil and gas industry. Although identified as a
key capability that E&P needs to leverage, only 13% of survey respondents
felt that their companies had mature analytics capabilities, and almost two-
thirds said that their companies would be investing in analytics over the next
years to close the gap.
Traditionally, the E&P industry-standard approach for data analysis has
been heavily leveraging mainstream spreadsheet-based tools and basic,
macro- or script-enabled workflow automation. With the new paradigm
of Big Data and the associated complexity as outlined previously in the
description of the 7 Vs, the traditional data analysis tools and techniques
quickly become suboptimal, due to intricate, nonlinear, multivariate, and
nonintuitive root-cause data relationships that affect decision-making.
At SPE Forum series event, “Next Generation of Smart Reservoir
Management: The Eminent Role of Big Data Analytics,” held in 2016 in
Dubai, UAE, the authors have concluded that sustainable transformation
of E&P businesses to fully harness the potential of Big Data requires scalable
data analytics with cognitive abilities, like massively distributed data mining
and machine and statistical learning, with little or no human supervision.
This is briefly addressed in the following sections. An efficient implementa-
tion of Big Data analytics can only be enabled by the innovative IT and data
management solutions that allow access to all data, all the time, and by all
stakeholders. This type of access increasingly being provided through the
implementation of flexible, open data management architectures such as