Page 137 - Intelligent Digital Oil And Gas Fields

CHAPTER FOUR

Components of Artificial

Intelligence and Data Analytics

Contents
4.1 Introduction 101
4.1.1 Artificial Intelligence: Overview of State of the Art in E&P 105
4.1.2 Data Analytics: Descriptive, Diagnostic, Predictive, Prescriptive, and Cognitive 108
4.1.3 Big Data in E&P: Concepts and Platforms 111
4.2 Intelligent Data Analytics and Visualization 115
4.2.1 Data Mining 115
4.2.2 Statistical and Machine Learning 117
4.2.3 Visualization and Interactivity 127
4.3 Applications to Digital Oil and Gas Fields 131
4.3.1 Machine Learning and Predictive Analytics 131
4.3.2 Data Mining, Multivariate, Root-Cause, and Performance Analysis 135
4.3.3 Event Diagnostics and Failure Analysis 139
4.3.4 Real-Time Analytics on Streaming Data 143
References 145
Further Reading 148

4.1 INTRODUCTION
As asset yields become harder to assess, extract, and forecast, oil and gas
operating companies and service providers must enable real-time decision-
making to better predict business outcomes that drive higher efficiencies and
utilization to achieve improved bottom-line results and profitability. With
the continued worldwide expansion of the digital oil field (DOF), the explo-
ration and production (E&P) industry is rapidly becoming an information-
and data-driven business.
If we accept the prediction that the DOF market will exceed $30 billion
by 2020 (Markets and Markets, 2015) along with exponential growth in vol-
ume and complexity of acquired data, the E&P industry needs to rapidly
adopt the new generation of digital transformation, technology, and pro-
cesses that include the following:

Intelligent Digital Oil and Gas Fields © 2018 Elsevier Inc. 101
https://doi.org/10.1016/B978-0-12-804642-5.00004-9 All rights reserved.