Cover
Title Page
Foreword
Foreword
1. WHAT ANALYTICS IS NOT
2. WHAT ANALYTICS ARE
Preface
Acknowledgments
CHAPTER 1: Introduction to Data‐Driven Concepts
1. INTRODUCTION
2. A DATA‐DRIVEN STUDY TIMELINE
3. IS IT INDUCTION OR DEDUCTION?
4. REFERENCES
CHAPTER 2: Data‐Driven Analytical Methods Used in E&P
1. INTRODUCTION
2. SOFT COMPUTING TECHNIQUES
3. RULES‐BASED METHODS
4. ARTIFICIAL INTELLIGENCE: MACHINE AND DEEP LEARNING
5. REFERENCES
CHAPTER 3: Advanced Geophysical and Petrophysical Methodologies
1. INTRODUCTION
2. ADVANCED GEOPHYSICAL METHODOLOGIES
3. ADVANCED PETROPHYSICAL METHODOLOGIES
4. REFERENCES
CHAPTER 4: Continuous Monitoring
1. INTRODUCTION
2. CONTINUOUS MONITORING IN THE RESERVOIR
3. MACHINE LEARNING TECHNIQUES FOR TEMPORAL DATA
4. SPATIOTEMPORAL PERSPECTIVES
5. TIME SERIES ANALYSIS
6. ADVANCED TIME SERIES PREDICTION
7. DIGITAL SIGNAL PROCESSING THEORY
8. HYDRAULIC FRACTURE MONITORING AND MAPPING
9. COMPLETIONS EVALUATION
10. RESERVOIR MONITORING: REAL‐TIME DATA QUALITY
11. DISTRIBUTED ACOUSTIC SENSING
12. DISTRIBUTED TEMPERATURE SENSING
13. CASE STUDY: TIME SERIES TO OPTIMIZE HYDRAULIC FRACTURE STRATEGY
14. REFERENCES
CHAPTER 5: Seismic Reservoir Characterization
1. INTRODUCTION
2. SEISMIC RESERVOIR CHARACTERIZATION: KEY PARAMETERS
3. MODULAR ARTIFICIAL NEURAL NETWORKS
4. WAVELET ANALYSIS
5. SPECTRAL DECOMPOSITION
6. FIRST ARRIVALS
7. NOISE SUPPRESSION
8. REFERENCES
CHAPTER 6: Seismic Attribute Analysis
1. INTRODUCTION
2. TYPES OF SEISMIC ATTRIBUTES
3. SEISMIC ATTRIBUTE WORKFLOWS
4. SEISMIC FACIES CLASSIFICATION
5. REFERENCES
CHAPTER 7: Geostatistics: Integrating Seismic and Petrophysical Data
1. INTRODUCTION
2. THE COVARIANCE AND THE VARIOGRAM
3. CASE STUDY: SPATIALLY PREDICTED MODEL OF ANISOTROPIC PERMEABILITY
4. KRIGING AND CO‐KRIGING
5. GEOSTATISTICAL INVERSION
6. KNOWLEDGE SYNTHESIS: BAYESIAN MAXIMUM ENTROPY (BME)
7. REFERENCES
CHAPTER 8: Artificial Intelligence: Machine and Deep Learning
1. INTRODUCTION
2. DATA MANAGEMENT
3. MACHINE LEARNING METHODOLOGIES
4. DEEP LEARNING TECHNIQUES
5. DEEP NEURAL NETWORK ARCHITECTURES
6. SEISMIC FEATURE IDENTIFICATION WORKFLOW
7. REFERENCES
CHAPTER 9: Case Studies: Deep Learning in E&P
1. INTRODUCTION
2. RESERVOIR CHARACTERIZATION
3. CASE STUDY: SEISMIC PROFILE ANALYSIS
4. CASE STUDY: ESTIMATED ULTIMATE RECOVERY
5. CASE STUDY: DEEP LEARNING APPLIED TO WELL DATA
6. CASE STUDY: GEOPHYSICAL FEATURE EXTRACTION: DEEP NEURAL NETWORKS
7. CASE STUDY: WELL LOG DATA‐DRIVEN EVALUATION FOR PETROPHYSICAL INSIGHTS
8. CASE STUDY: FUNCTIONAL DATA ANALYSIS IN RESERVOIR MANAGEMENT
9. REFERENCES
Glossary
About the Authors
Index
End User License Agreement

List of Illustrations

Chapter 02
1. Figure 2.1 Multidisciplinary nature of soft computing technologies
2. Figure 2.2 Analytical workflow demonstrating an ensemble model
3. Figure 2.3 Simplest expression of a neural network
4. Figure 2.4 Perceptron architecture for a neural network
5. Figure 2.5 Hidden layer within the neural network architecture
6. Figure 2.6 Gradient descent solves linear regression
Chapter 03
1. Figure 3.1 A typical spectral window of 4500 samples from a single component channel of passive seismic data lasting 9 seconds from a sea‐floor seismic array. The signal to noise ratio is about 1:1. Dominant frequencies include 60 Hz mains AC and its harmonics, and much lower frequencies from nearby drilling.
2. Figure 3.2 Petrophysical business studies in a data science context
3. Figure 3.3 A statistical study of well log data indicates that the elastic impedance at the formation level across 180 wells falls into two clusters. Closer inspection of the data revealed that a milli‐ prefix had persisted for some wells. This is a data governance issue to be addressed before data can be used for analysis.
4. Figure 3.4 Average formation density and its standard deviation for the BRENT group (Broom blue/circle, Rannoch pink/X, Etive brown/square, Ness purple/+, Tarbert red/*). The more heterogeneous Etive formation contrasts with the homogeneous sandstone of the Broom. Of interest are the outlier wells for each group and the diversity in variability across the Tarbert.
5. Figure 3.5 Coal measures determined in a borehole in the Taranaki Basin, NE New Zealand, were used to create a wavelet‐based template that was applied to the boreholes for the rest of the basin with great success. Rahman (2015) reports occasional false positives pointing to a need to tune the technique further, but also reports many candidate features that had not been interpreted as coal but occurred in interpreted formations that were associated with coal.
Chapter 04
1. Figure 4.1 Original time series convolved additive model
2. Figure 4.2 Decomposition panels illustrating trends and seasonal cycles
3. Figure 4.3 Time series patterns enable a hydrocarbon production gap analysis
4. Figure 4.4 Gap analysis for a typical well
5. Figure 4.5 Data QC workflows
6. Figure 4.6 Data QC solution platform
7. Figure 4.7 Surface plot of temperature DTS data
8. Figure 4.8 Surface plot illustrating temperature gradients across the entire fractured wellbore
9. Figure 4.9 Sankey diagram for the cumulative gas production
10. Figure 4.10 Correlation matrix
11. Figure 4.11 Bubble plot
12. Figure 4.12 Network diagram
13. Figure 4.13 SOM diagram—Qg100
14. Figure 4.14 SOM diagram—Bulk modulus
15. Figure 4.15 SOM diagram—Fracture stages
16. Figure 4.16 SOM diagram—Proppant volume
Chapter 05
1. Figure 5.1 Multivariate, multivariant, multidimensional, and stochastic seismic characteristics
2. Figure 5.2 Traditional seismic fault interpretation
3. Figure 5.3 Seismic reservoir characterization data‐driven methodology
4. Figure 5.4 Principal component analysis workflow
5. Figure 5.5 Line plot showing seismic trace data post F‐K transform
6. Figure 5.6 Summary of the decomposition analysis illustrating the wavelet coefficients
7. Figure 5.7 Detail coefficient plot for all levels with independent scaling
8. Figure 5.8 Detail coefficient plot with the top three levels scaled uniformly
9. Figure 5.9 Detail coefficient plot using the Donoho and Johnstone algorithm
10. Figure 5.10 Multiresolution approximation plot showing restorations of the input signal
11. Figure 5.11 Multiresolution approximation plot of a particular level
12. Figure 5.12 Multiresolution approximation plot for level 10
13. Figure 5.13 Multiresolution decomposition plot for the seismic data
14. Figure 5.14 Wavelet scalogram of the seismic trace
15. Figure 5.15 Wavelet scalogram showing most energy at level 8
16. Figure 5.16 Line plot showing seismic trace data post wavelet smoothing
17. Figure 5.17 True surface g(t, x)
18. Figure 5.18 Noisy observations Z(t, x)
19. Figure 5.19 Estimated smooth surface
20. Figure 5.20 Shot record with ground roll before applying noise suppression algorithm
21. Figure 5.21 Shot record after applying noise suppression algorithm
22. Figure 5.22 The attenuated noise
Chapter 06
1. Figure 6.1 SEMMA process for data‐driven analysis
2. Figure 6.2 Classification in a seismic facies interpretation iterative cycle
3. Figure 6.3 Seismic facies classification workflow
4. Figure 6.4 Dendrogram generated by a hierarchical clustering algorithm
5. Figure 6.5 Constellation plot
6. Figure 6.6 k‐means cluster visualization
7. Figure 6.7 Clusters depicted with contours for the normal densities
8. Figure 6.8 Ternary plot explaining the cluster probabilities for each seismic observation
9. Figure 6.9 Principal components on correlations report
10. Figure 6.10 Seismic classification set of clusters from a business perspective
Chapter 07
1. Figure 7.1 Geostatistics and earth modeling
2. Figure 7.2 Basic assumption of geostatistics
3. Figure 7.3 Trend surface analysis
4. Figure 7.4 Scatter plot of permeability observations at discrete locations across the reservoir
5. Figure 7.5 Histogram detailing lag classes and the trends in the measurement
6. Figure 7.6 Scatter plot of permeability data detrended residuals
7. Figure 7.7 Semivariograms for varying angles θ = 0° to 45°
8. Figure 7.8 Semivariograms for changing angles θ = 60° to 105°
9. Figure 7.9 Semivariograms for changing angles θ = 120° to 165°
10. Figure 7.10 Boxplot of the square root difference cloud
11. Figure 7.11 Semivariograms for angles θ = 0° and θ = 90°
12. Figure 7.12 Semivariogram for permeability based on the selected Gaussian‐Gaussian model
13. Figure 7.13 Kriging prediction for permeability values across reservoir using selected model
14. Figure 7.14 Semivariogram for permeability based on the exponential model
15. Figure 7.15 Kriging prediction for permeability values across reservoir using exponential model
16. Figure 7.16 Knowledge Synthesis foundation
17. Figure 7.17 Knowledge Synthesis implementing Bayesian maximum entropy methodology
Chapter 08
1. Figure 8.1 Machine learning taxonomy
2. Figure 8.2 Concrete and colorful property space concept
3. Figure 8.3 Red, green, and blue (RGB) color cube
4. Figure 8.4 Typical open workflow based on derived seismic attributes
5. Figure 8.5 Supervised and unsupervised data mining techniques in deep learning
6. Figure 8.6 Deep forward neural network
7. Figure 8.7 Convolutional deep neural network
8. Figure 8.8 5 × 5 seismic image convolved with a 3 × 3 matrix to generate another 3 × 3 matrix of pixel values
9. Figure 8.9 Max‐pooling function applied to demarcate a spatial neighborhood
10. Figure 8.10 Recurrent deep neural network
11. Figure 8.11 Stacked denoising autoencoder details
12. Figure 8.12 Seismic features generated for deep learning methodology
13. Figure 8.13 Stacked denoising autoencoder
14. Figure 8.14 Convert pixels of interest to grayscale
15. Figure 8.15 Break grayscale images into thousands of overlapping patches
16. Figure 8.16 Illustration of overlapping patches in an image
17. Figure 8.17 Patch vectorization rearranges pixels as a row vector in lexicographic order
18. Figure 8.18 Example of a dictionary that has 12 atoms
19. Figure 8.19 Schematic of a stacked autoencoder. Outputs are set equal to input patterns and the learned weights of the output layer from the dictionary
Chapter 09
1. Figure 9.1 Reservoir characteristic workflows using DL technologies
2. Figure 9.2 Classification as applied to the interpretation of seismic facies (Duda et al., 2000)
3. Figure 9.3 High‐resolution seismic images
4. Figure 9.4 Sample 20 × 20 pixel patches created from high‐resolution seismic images
5. Figure 9.5 Four clusters of patches overlaid onto a single original inline 2D seismic image
6. Figure 9.6 Dictionary defines new direct hydrocarbon indicators (DHIs)
7. Figure 9.7 Single‐layer autoencoder trained by stochastic gradient descent optimization
8. Figure 9.8 Sample of the dictionary images
9. Figure 9.9 Exploratory data analysis visualizing the histograms of key variables
10. Figure 9.10 RBM architecture implemented to predict output
11. Figure 9.11 Flat‐spot is a seismic attribute anomaly depicted as a horizontal reflector
12. Figure 9.12 Simple neural architecture for flat‐spot identification
13. Figure 9.13 Introduction of layer A focused on a small time–space window
14. Figure 9.14 Layer of neurons B adds a convolutional layer for focused analysis
15. Figure 9.15 A max‐pooling layer helps to identify seismic features
16. Figure 9.16 Workflow details the acquisition of raw hard data from logging tools that are used to determine petrophysical properties
17. Figure 9.17 Functional data analysis
18. Figure 9.18 Cluster analysis to segment the well portfolio in a brownfield

Wiley & SAS Business Series

The Wiley & SAS Business Series presents books that help senior‐level managers with their critical management decisions.

Titles in the Wiley & SAS Business Series include:

Analytics: The Agile Way by Phil Simon
Analytics in a Big Data World: The Essential Guide to Data Science and Its Applications by Bart Baesens
A Practical Guide to Analytics for Governments: Using Big Data for Good by Marie Lowman
Bank Fraud: Using Technology to Combat Losses by Revathi Subramanian
Big Data Analytics: Turning Big Data into Big Money by Frank Ohlhorst
Big Data, Big Innovation: Enabling Competitive Differentiation through Business Analytics by Evan Stubbs
Business Analytics for Customer Intelligence by Gert Laursen
Business Intelligence Applied: Implementing an Effective Information and Communications Technology Infrastructure by Michael Gendron
Business Intelligence and the Cloud: Strategic Implementation Guide by Michael S. Gendron
Business Transformation: A Roadmap for Maximizing Organizational Insights by Aiman Zeid
Connecting Organizational Silos: Taking Knowledge Flow Management to the Next Level with Social Media by Frank Leistner
Data‐Driven Healthcare: How Analytics and BI are Transforming the Industry by Laura Madsen
Delivering Business Analytics: Practical Guidelines for Best Practice by Evan Stubbs
Demand‐Driven Forecasting: A Structured Approach to Forecasting, Second Edition by Charles Chase
Demand‐Driven Inventory Optimization and Replenishment: Creating a More Efficient Supply Chain by Robert A. Davis
Developing Human Capital: Using Analytics to Plan and Optimize Your Learning and Development Investments by Gene Pease, Barbara Beresford, and Lew Walker
The Executive's Guide to Enterprise Social Media Strategy: How Social Networks Are Radically Transforming Your Business by David Thomas and Mike Barlow
Economic and Business Forecasting: Analyzing and Interpreting Econometric Results by John Silvia, Azhar Iqbal, Kaylyn Swankoski, Sarah Watt, and Sam Bullard
Economic Modeling in the Post–Great Recession Era: Incomplete Data, Imperfect Markets by John Silvia, Azhar Iqbal, and Sarah Watt House
Enhance Oil and Gas Exploration with Data‐Driven Geophysical and Petrophysical Models by Keith Holdaway and Duncan Irving
Foreign Currency Financial Reporting from Euros to Yen to Yuan: A Guide to Fundamental Concepts and Practical Applications by Robert Rowan
Harness Oil and Gas Big Data with Analytics: Optimize Exploration and Production with Data‐Driven Models by Keith Holdaway
Health Analytics: Gaining the Insights to Transform Health Care by Jason Burke
Heuristics in Analytics: A Practical Perspective of What Influences Our Analytical World by Carlos Andre Reis Pinheiro and Fiona McNeill
Human Capital Analytics: How to Harness the Potential of Your Organization's Greatest Asset by Gene Pease, Boyce Byerly, and Jac Fitz‐enz
Implement, Improve and Expand Your Statewide Longitudinal Data System: Creating a Culture of Data in Education by Jamie McQuiggan and Armistead Sapp
Intelligent Credit Scoring: Building and Implementing Better Credit Risk Scorecards, Second Edition by Naeem Siddiqi
Killer Analytics: Top 20 Metrics Missing from Your Balance Sheet by Mark Brown
Machine Learning for Marketers: Hold the Math by Jim Sterne
On‐Camera Coach: Tools and Techniques for Business Professionals in a Video‐Driven World by Karin Reed
Predictive Analytics for Human Resources by Jac Fitz‐enz and John Mattox II
Predictive Business Analytics: Forward‐Looking Capabilities to Improve Business Performance by Lawrence Maisel and Gary Cokins
Profit Driven Business Analytics: A Practitioner's Guide to Transforming Big Data into Added Value by Wouter Verbeke, Cristian Bravo, and Bart Baesens
Retail Analytics: The Secret Weapon by Emmett Cox
Social Network Analysis in Telecommunications by Carlos Andre Reis Pinheiro
Statistical Thinking: Improving Business Performance, Second Edition by Roger W. Hoerl and Ronald D. Snee
Strategies in Biomedical Data Science: Driving Force for Innovation by Jay Etchings
Style and Statistics: The Art of Retail Analytics by Brittany Bullard
Taming the Big Data Tidal Wave: Finding Opportunities in Huge Data Streams with Advanced Analytics by Bill Franks
Too Big to Ignore: The Business Case for Big Data by Phil Simon
The Analytic Hospitality Executive by Kelly A. McGuire
The Value of Business Analytics: Identifying the Path to Profitability by Evan Stubbs
The Visual Organization: Data Visualization, Big Data, and the Quest for Better Decisions by Phil Simon
Using Big Data Analytics: Turning Big Data into Big Money by Jared Dean
Win with Advanced Business Analytics: Creating Business Value from Your Data by Jean Paul Isson and Jesse Harriott

For more information on any of the above titles, please visit www.wiley.com.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per‐copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750‐8400, fax (978) 646‐8600, or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748‐6011, fax (201) 748‐6008, or online at www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762‐2974, outside the United States at (317) 572‐3993, or fax (317) 572‐4002.

Wiley publishes in a variety of print and electronic formats and by print‐on‐demand. Some material included with standard print versions of this book may not be included in e‐books or in print‐on‐demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com. For more information about Wiley products, visit www.wiley.com.

Library of Congress Cataloging‐in‐Publication Data is available:

Names: Holdaway, Keith R., author. | Irving, Duncan H. B., 1971– author.
Title: Enhance oil & gas exploration with data‐driven geophysical and petrophysical models / by Keith R. Holdaway, Duncan H.B. Irving.
Other titles: Enhance oil and gas exploration with data‐driven geophysical and petrophysical models
Description: Hoboken, New Jersey : Wiley, 2018. | Includes bibliographical references and index. |
Identifiers: LCCN 2017027921 (print) | LCCN 2017040698 (ebook) | ISBN 9781119302599 (pdf) | ISBN 9781119302582 (epub) | ISBN 9781119215103 (hardback)
Subjects: LCSH: Petroleum—Prospecting—Mathematics. | Prospecting—Geophysical methods—Mathematics. | Petroleum—Geology—Mathemaical models. | BISAC: BUSINESS & ECONOMICS / Industries / Energy Industries.
Classification: LCC TN271.P4 (ebook) | LCC TN271.P4 H653 2018 (print) | DDC 622/.1828—dc23
LC record available at https://lccn.loc.gov/2017027921

Cover Design: Wiley
Cover Image: © naqiewei/Getty Images

Foreword

I vividly remember the first time I met Keith Holdaway. It was 14 years ago, and he was standing in the front row of an analytics conference. He cut a distinctive profile as he challenged the speaker at the podium, asserting quite stubbornly that the oil and gas industry could realize huge returns by using a more data‐driven approach that exploited the full potential of analytics. As a young man (or so I thought of myself at the time), I had been tasked with selling analytical software to upstream oil and gas companies. Coming from a technology background, I realized that this gentleman was the guide I was looking for and made a mental note to seek him out at the cocktail hour.

Back then, in 1989, the digital oilfield was the topic of the day, promising impressive returns. As the industry embraced the concept more fully over the next decade, I observed companies making significant investments in specific data solutions to automate and solve a broad range of problems. Thought leaders eagerly embraced the application of data‐driven analytics, but the adoption was not necessarily as widespread as one would have thought. Scattershot adoption created its issues, with companies sometimes running hundreds of disparate applications and ending up with silos of data across their organizations. The promise remained.

Fast forward to 2014 and Keith's first book, Harness Oil and Gas Big Data with Analytics, which arrived just before crude plunged to historic lows. In retrospect his book seems almost prescient as the industry's enthusiasm for data‐driven analytics has been driven in part by the potential to generate greater value from its assets in the face of a much lower price per barrel. Many of the leading players—and several influential thought leaders among smaller oil companies—have made substantial investments in this area, and there is more to come. Increasingly, I am contacted by clients looking for data scientists, asking for training, and seeking guidance on how best to implement advanced analytics programs. We often point them to Keith's book, among other resources at SAS and elsewhere, to help them validate the best path forward.

Hence the genesis of this new book. Interest in his first book has been consistent enough that colleagues implored Keith to write a second volume: a more particular text that digs deeper into applying data‐driven approaches across the exploration sector. Keith and his colleague, Dr. Duncan Irving, have written an invaluable book, exploring the data‐driven methodologies in the disciplines of geophysics and petrophysics. And the timing is right. We are witnessing an unprecedented convergence of big data and cloud technology with massive increases in computing power at a time when a climate of low prices has made driving efficiencies an absolute requirement. Add to that the influx of technology‐attuned Millennials into the workforce, and oil and gas practitioners are on the verge of a new era of opportunity to transform their business.

I have no doubt that this volume will be a valuable addition to the growing body of resources focused on this exciting area. Over years of working at the nexus of energy and technology, Keith has become a mentor and friend. His colleague is a globally recognized geophysicist working in the field of data analytics and brings innovative ideas to the evolving science of data‐driven and soft‐computing technologies. This new and important book is the result of years of deep work in this area and a real passion for the topic, approached with the same determination I saw at the front of that conference room many years ago. I am honored to introduce this book: Enhance Oil and Gas Exploration with Data‐Driven Geophysical and Petrophysical Models.

Ross Graham,
Director, O&G Americas
Calgary, June 2017

Foreword

The oilfield is one of the most data‐rich industries in the world, and concerning real information (as opposed to virtual data generated by the web and other virtual environments) can lay claim to the most data intensive industry. Most organizations, if they are honest with themselves, rarely capitalize on the potential of analytics and ‘big data.’ The authors of this book address the most common pitfalls that beset analytics and provide a comprehensive framework and roadmap, from the exploration and production perspective, to achieve the real goal of analytics—simplifying, expediting, or making possible the translation of data into profitable and sustainable outcomes.

To unleash the power of analytics, one must first understand what they are and are not. Analytics are data‐centric processes that, if designed and executed properly, will lead to insights and outcomes. Each aspect of the process must receive due diligence, and the focus of the endeavor should always be to add value to the organization.

WHAT ANALYTICS IS NOT

The most common mistake when understanding analytics is to confuse the sizzle with the steak—that is to conflate the perception of a thing with the substance of the thing. Many managers and even technical professionals accept the misconception that analytics is the collation and visualization of data using colorful charts and graphs. This is not only incorrect, but there is a tacit danger in this assumption because it can significantly limit future analytic endeavors that do not, per se, yield an attractive visual. It must be understood, therefore, that dashboards and reports are one of many results of analytics and, while they are the most visible, they may not be the most valuable.

WHAT ANALYTICS ARE

Analytics are multi‐step processes which transform data from one or more sources into information which leads to changes in actions and behaviors; and, if an organization is unwilling to do either, investment in analytics should be reconsidered. This book, more than any other before it, details a simple, yet robust, approach to developing an analytics plan that will lead to success. Though analytics methodologies vary depending on query most processes should contain at least the following:

Data Modeling. Analytics planning should ensure, within practical limits, that necessary and sufficient data are identified beforehand.
Data Gathering with a focus on quality. Identification and management of adverse data are often far more resource intensive and problematic than data that is missing. Acquiring real data often involves rigorous technical and contract specifications that include detailed definitions of data properties.
Data Management—how data will be transferred, stored, secured, transformed, and distributed.
Analysis—Understanding which analytical methods are most appropriate based on types of data and questions asked as well as the speed and accuracy of the desired results.
Communication—Determining the most efficient and influential modes in which to communicate data to those who should, or could, consume it—whether it is formal reports, presentations, email, social media, audiovisual, or combination of these and other forms.
Management of Change. Perhaps the most important, yet sadly overlooked, part of an analytics project involves: identifying, before work begins, who all relevant stakeholder (or customers) are, clearly documenting their needs, and agreeing in advance on if, or how, changes to process might occur based on the results of analyses.

Nathan Zenero
President,
Verion Applied Technologies

Preface

Our motivation for writing this book comes from the professional curiosity and experience we have accumulated over recent years in the Oil and Gas industry. We have noted and continue to witness the struggles between geoscientists and their multiple spatial and temporal datasets. Traditional interpretation can provide certain answers based on Newtonian physics and the fundamental laws of nature, but with so much data being amassed with sensors in this digital age, it is necessary to marry deterministic interpretation with data‐driven workflows and soft‐computing models.

Owing to the cyclical nature of the Oil and Gas industry, we have seen historically depressed crude prices since 2015. This last downturn, like previous historical downturns, shook the industry to the point of an overreaction: people losing their livelihoods, reduction in OPEX, and cancellation of projects, particularly in exploration. It is at these transition points that oil and gas companies seek more efficient work processes and best practices. This invariably results in the adoption of technologies not necessarily new in other industries. Today we see more adoption of soft‐computing and data‐driven analytics to complement the traditional interpretation.

Given these cyclical‐downturn scenarios, we ask ourselves, being in the trough of a current downturn: What's happening in the Oil and Gas industry today?

We are aware of the dramatic drop in crude oil prices that is a driver behind the industry's march toward adopting new technologies such as analytical and soft‐computing workflows. Oil and gas companies realize the climb from the bottom of the cycle is a slow process and has many global and local influences. Too much supply and weak global demand play into a dynamic scenario.

Oil and gas companies are currently contemplating serious near‐term investments to develop global assets, but it behooves the industry to move gingerly. We shall witness an inexorably slow increase in oil prices, with global supply bound by the reduction in reserve development projects over the past few years.

Many talented engineers have left the industry, and the internal organizational vagaries, coupled with inflexible and complex systems, processes, and attitudes could put the breaks on any innovative and evolving methodologies and best practices. IOCs and NOCs are looking seriously at a digitization environment using advanced analytics for the new daily workflows. Service companies, analytics vendors, and in‐house capabilities are emerging to address these needs. This will enable oil and gas companies to weather current and future industry downturns.

We see this book as a contribution to enabling upstream geoscientists in data‐driven analytics in geophysics and petrophysics. We hope it serves to bring together the practitioners of conventional upstream computing workflows with the new breed of data scientist and analyst and generate overlap and common ground so they can understand each other's perspectives, approaches, and role in this new computing landscape.