oil and gas resources library
Novi Cube Optimization Featured in JPT
Novi’s application of machine learning to optimize cube developments has been featured in JPT! The article highlights recent Novi results showing that the cube is not dead — smart operators can learn from the mistakes of others. Each cube leaves a signature–perfect for machine learning to digest.
Novi Presents :: Oil & Gas Cube Development Webinar Replay
In this open discussion, Novi experts take a dive into:
What levers do oil and gas operators have to make it work today?
What are the challenges in planning a cube?
What does the future of cube planning look like?
Check out the webinar replay available for a limited time! You can also gain more insight on this hot topic in one of our latest blogs on Oil & Gas Cube Development: How Dense is Too Dense?
URTeC 2021: Are unconventional well performance gains exhausted? Investigating the drivers of year-over-year production improvements across the major US unconventional plays using machine learning (ID 5263)
What has driven the unconventional productivity gains over the past decade, and can we expect continued performance improvements? We use our machine learning models to analyze this question across multiple basins.
URTeC 2021: Quantifying the Diminishing Impact of Completions Over Time Across the Bakken, Eagle Ford, and Wolfcamp Using a Multi-Target Machine Learning Model and SHAP Values (ID 5633)
In this paper we use our models to evaluate how the impact of completions varies over the life of a well, across multiple basins. Using an estimate of completions uplift at a single point in time can result in inaccurate forecasts and economics.
URTeC 2021: Machine Learning Methods in the Williston: A Case Study in Productivity Decay and the Implications for Inventory Exhaustion (ID 5303)
What does remaining inventory look like in the Bakken? This study examines how rock quality of remaining inventory varies around the basin, and what that means for future breakevens and supply forecasts.
The Power of Novi Data Engine
After six years of building machine learning models for unconventional development, we have learned a HUGE amount about what it takes to build analytics-ready datasets. It’s not easy: messy tables, multiple sources, industry-specific challenges, and complicated transformations to get the data right.
Now, we are putting all of our experience and (and custom data-processing algorithms) in your hands with Data Engine. With our V1 release, we are starting at the beginning of the process: uploading, mapping, and formatting the data sources.
Data Engine V1 provides immediate benefit by shortening time-to-insights and putting more control of the data in YOUR hands. But beyond that, it represents the first step towards Novi as a full self-service platform.
Novi Presents at SPE Dallas Section Study Group: Machine Learning 101 in the Midland Basin
Novi Technical Advisor Ted Cross will be presenting at the SPE Dallas Section Study Group, covering the basics of machine learning: when to use machine learning, data preparation, measuring performance, how to explain model results, and leveraging the model for optimization scenarios. The presentation will then review applications of these concepts in the Midland Basin using a public-data model, looking at completions sensitivities, geology-spacing interactions, and asset development optimization case studies.
AAPG 2020: Before the Basin Model: Using ML & Formation Tops to Understand the Effect of Burial History (ID 4694)
What is the minimum amount of geological data required to build a predictive model? Traditional basin modeling approaches utilize expensive datasets and significant time expenditure to understand and predict fluid distribution around the basin. Machine learning offers an empirical method to generate similar predictions. In contrast to the forward-modeling method of traditional basin modeling, machine learning looks for implicit relationships between training and target variables. In this study, we investigate whether a machine learning model fed only geologic tops as input geological data can accurately predict well performance in the Bakken-Three Forks play of the Williston Basin.
AAPG 2020: Tailoring Completions to Geology: A Machine Learning Approach (ID 4686)
Not all rocks should receive the same completion. It’s a simple concept, but it can be tough to back up the theory with statistical evidence. Machine learning provides a powerful tool to analyze the interactions between completions design choices and local geology. In this talk, we analyze interactions between proppant, fluid loading, stage spacing, and geology. We then show that per-well economics can be improved by over $1MM NPV by taking a tailored approach. SHAP values (like in the image above) show what the model is thinking: what exact combinations of geology & spacing are driving these interactions.
AAPG 2020: Prior Well Depletion or Interwell Spacing? Isolating the Causes of Spacing Degradation in the Williston Basin (ID 4697)
Understanding the impact of interwell spacing is tough enough before introducing the impact of parent-child depletion. When you add in the fact that both have a strong correlation with geology—especially in the Bakken, where much of the early producers and early downspacing tests were in the best rock–well, then things can really get hairy. In this talk, we investigate a variety of ways of measuring parent-child and spacing impact, including days online, number of parents within a radius, parent production, and distance to neighbors.
Novi Water Modeling featured in JPT September Issue
Novi’s URTeC 2020 paper, Predicting Water Production in the Williston Basin Using a Machine-Learning Model has been featured in the September issue of JPT.
The article highlights the many ways that the potent combination of machine learning models plus “explainability” datasets like SHAP values (from Game Theory!) can add value for operators, midstream companies, and local stakeholders. Digital oilfield software developed for oil forecasting can be easily adapted for water forecasting!
URTeC 2020: Deriving Time-Dependent Scaling Factors for Completions Parameters in the Williston Basin using a Multi-Target Machine Learning Model and Shap Values (ID 3103)
Applying a single scaling factor over the life of the well for a completions variable will give you an inaccurate production profile — and inaccurate economics! Quantifying the time-variant impact of completions decisions is tough for traditional methods. However, training a machine learning model to predict a time series of production, combined with Shapley values, provides a powerful tool to investigate when completions are most impactful.
URTeC 2020: Predicting Water Production in the Williston Basin using a Machine Learning Model (ID 2756)
Like it or not, most companies don’t spend nearly as much time analyzing water production as oil (and who can blame them!?!). Fortunately, machine learning techniques developed for forecasting oil production can be adapted for water production — it’s part of our standard model delivery.
URTeC 2020: Decomposition of Publicly Reported Combined Hydrocarbon Streams using Machine Learning in the Montney and Duvernay (ID 2795)
Canadian operators have an information problem: in the public data, “gas” wells don’t have to report their liquids content. Novi has developed a solution to estimate CGR that’s twice as accurate as the traditional methods.
URTeC 2020: GeoSHAP: A Novel Method of Deriving Rock Quality Index from Machine Learning Models and Principal Components Analysis (ID 2743)
Since Novi was founded, customers, partners, and prospects have been asking us if machine learning could be used to extract a subsurface model directly from logs. The answer is yes.
URTeC 2020: Evaluating the Impact of Precision Targeting on Production in the Midland Basin using Machine Learning Algorithms (ID 3062)
With the right dataset and right machine learning model, you can get high-resolution answers on fine-tuning your target selection — potentially increasing production by over 10% at zero cost.
Using Prediction Engine to Evaluate Whiting’s Undrilled Acreage
What are the best remaining acreage blocks in Whiting’s North Dakota portfolio? With Prediction Engine, you can quickly and easily lay out well sticks and get predictions for different designs. Here, we walk you through how to configure & generate economics across the acreage.
Impact of delay-and-infill strategy on recoveries and returns :: Parsley – Jagged Peak acreage
At $30 oil, marginal zones become uneconomic. What is the impact on EURs and returns if those zones are infilled two years in the future? In this video, we take a look at the impact of delays on Bone Springs Third Sand and Wolfcamp A, B, and C zones in Parsley’s newly-acquired legacy Jagged Peak acreage in the southeastern Delaware Basin.
Using game theory and confidence interval outputs from a machine learning model to risk adjust oil well returns
In this video we walk through how Novi Shapley data and confidence interval data can be used to gain confidence in machine learning model predictions and handicap possible well performance using model confidence as a proxy for risk. You can review the blog post here.
Iterating and optimizing completion designs across NBL’s entire Delaware DUC inventory in minutes
In this Novi product video, Novi Technical Advisor John Chaplin walks through how a user can create Novi Prediction Engine Outputs, using NBL’s DUC inventory as an example, and publish these to Novi Cloud for analysis. To see the full description of the inputs and how we put this together read the blog post here.
Given sub-$30 wti strip, how can NBL maximize returns on their DUC inventory?
In this Novi product video, Novi Technical Advisor John Chaplin walks through an analysis of NBL’s Delaware DUC inventory and suggests changes in completion design in this sub $30 WTI price environment based on Novi Prediction Engine Outputs published to Novi Cloud. To see the full description of the inputs and how we put this together read the blog post here.
Changing plans on a dime: analyzing QEP’s Midland asset given sub-$30 WTI strip with Novi Prediction Engine
In this Novi product video, Novi Technical Advisor John Chaplin walks through and analysis of QEP’s Midland asset and suggests changes in capital allocation in this sub $30 WTI price environment based on Novi Prediction Engine Outputs published to Novi Cloud. To see the full description of the inputs and how we put this together read the blog post here.
Deriving a regional subsurface model in a day from logs and tops – Novi Subsurface data extraction workflow
This video goes into detail on Novi’s subsurface model for the Williston Basin in North Dakota and is incorporated in a Novi blog post on this same subject. Novi algorithms analyzed & extracted log data, conducted a principal component analysis, and used that information for model training. When combined with SHAP values, this workflow becomes a powerful way to generate rock quality maps using machine learning.
AAPG Explorer article “Is the Next Oil Production Breakthrough Already Here?” featuring Novi President Jon Ludwig
Just prior to URTeC 2019, Barry Friedman of AAPG Explorer asked Novi President Jon Ludwig some important questions about the future of digital oilfield adoption – particularly when would rapid adoption start to happen across the industry.
Prediction Engine Version 2 Software Demonstration
A live video demonstration of Novi’s Prediction Engine version 2 software. Quickly create robust economic scenarios when valuing acreage, searching for the best development scenario given capital constraints, or understanding how Novi’s models interpret subsurface, stacking, spacing and stimulation intensity scenarios.
SHELL iSHALE® COLLABORATES WITH NOVI
Shell’s technology collaborations with Novi Labs LLC (“Novi”) promise to deliver smarter and more efficient wells in not too distant future.
“We are collaborating with our strategic partners to progress a technological step change in Shell’s shale assets and develop a shale field of the future.”
Novi / Range Resources SPE-191796-18ERM-MS: Integrating Big Data Analytics Into Development Planning Optimization
This paper reviews several Big Data analytical initiatives in the Marcellus Shale. We describe how application of Big Data technology evolved, share challenges and benefits derived from Big Data analytical processes, and discuss lessons learned. We present an overview of Big Data methods employed, show how we integrated results with economic analyses to guide field development, and summarize the significant impact on development economics.
Co-Develop vs. Infill – A Complex Question
Should all formations be developed at the same time (co-develop) or, should an Operator develop first tier formation targets first, then infill secondary formations later? This video demonstrates how Novi Data consumed through Novi Cloud can help answer this complex question, and define the economic loss of infilling vs. co-development.
Barclays Bank "Frac to the Future" Analyst Report (rel: January 2020)
Read Barclays Bank report “Frac to the Future” for some great insight on the Digital Oilfield technology marketplace. In depth analysis of the “Big Three” (Microsoft, Google, AWS) cloud providers, smaller companies such as Novi, and the legacy software providers to the industry (HAL, SLB).
And, Novi has a nice profile in there, and a lot of input into the report straight to the Barclays guys.
Evaluating PV10 acquisition scenarios using Novi Cloud outputs in Spotfire
Novi Founder and President Jon Ludwig walks through how to use a business intelligence tool like Spotfire to connect to Novi Cloud and consume Novi Prediction Engine outputs. See how Novi’s data can be used to instantly calculate IRR, NPV, PVI, cash payback and cashflow from multiple possible development plans in a single pane.
Using Novi Prediction Engine to evaluate Parsley Energy’s acquisition of Jagged Peak
In this video, Novi Technical Advisor Ted Cross will walk through a demonstration of Novi’s Prediction Engine software, just released on 02-February, 2020. See how Novi Prediction Engine will crush the amount of time it takes engineers, geoscientists, or financial analysts to do a very robust analysis of a play, both before its acquired, and after.
Novi AAPG ACE 2019 Paper – Production and Subsurface Machine Learning Model for Predicting Hydrocarbon Recovery
Download our AAPG ACE 2019 technical presentation! Presented by Novi’s chief geophysicist Kiran Sathaye, the presentation focuses on building a predictive #OilAndGas recovery model based on public data only, including an automated framework to derive a subsurface model directly from electronic well logs.
Picking the Best Statistical Approach for Modeling Hydrocarbon Systems
Geologic, reservoir, and production datasets present a vast array of variables when analyzing oil & gas recovery. What is the best approach to statistical modeling of highly complex hydrocarbon systems? This report compares two approaches: linear multi-variant analysis and a tree-based machine learning approach.
5 Ways to Prevent Over-Engineered and Overpriced Wells
Operators execute a balancing act when optimizing spacing, stacking, completion design and length parameters. However, type curves often result in well designs that are either over or under-engineered; resulting in poor deployment of capital and constrained recovery rates. In this report, we present the top 5 mistakes to avoid for well engineering.