The Novi AI Roundup

How is the Lower 48 continuing to produce more with less? In this episode of The Novi AI Roundup, we explore how technology and operational efficiency are reshaping the outlook for U.S. unconventional development. Drawing from Ted Cross’s presentation “Key Trends & Emerging Plays Across Lower 48”, we examine how fewer rigs are still supporting strong production, why longer laterals and better drilling efficiency are improving well performance, and how secondary targets and emerging plays are opening new opportunities across mature basins.This podcast episode is based on the presentation “Key Trends & Emerging Plays Across Lower 48”, written by Ted Cross. Download the full presentation here.

How should machine learning account for the impact of parent-well depletion on future production? In this episode of The Novi AI Roundup, we explore how different depletion features can change the way ML models forecast parent–child interactions. Drawing from the URTeC 2026 paper “Designing Depletion Features for Parent–Child Modeling: Comparing Time, Distance, and Volume–Distance Formulations in Machine-Learning Workflows,” we examine why distance-weighted depletion can better capture the relationship between parent production, proximity, and child-well performance, and what this means for forecasting and infill development planning.This podcast episode is based on the technical paper “Designing Depletion Features for Parent–Child Modeling: Comparing Time, Distance, and Volume–Distance Formulations in Machine-Learning Workflows,” authors: B. Davis, A. Qualls, D. Niederhut, and K. Sathaye. The paper was presented at URTeC 2026. Download the full paper here.

How does well spacing influence water production in the Midland Basin? In this episode of The Novi AI Roundup, we explore how machine learning can help quantify the relationship between spacing and water production. Drawing from the URTeC 2026 paper “Well Spacing Impacts Water Production in the Permian Basin,” we examine why spacing parameters and parent well history are important predictors of water cut, how the timing of child well completions can affect water production, and what these insights mean for water-aware development decisions.This podcast episode is based on the technical paper “Well Spacing Impacts Water Production in the Permian Basin,” authors: F. Male, R. Dommisse, and K. Sathaye. The paper was presented at URTeC 2026. Download the full paper here.

How do you accurately forecast wells in increasingly crowded developments? In this episode of The Novi AI Roundup, we explore how machine-learning-ready metrics can better capture the complex interactions between wells in ultra-tight infill developments. Drawing from the technical paper “Making an Impact: Machine-Learning-Ready Metrics for Ultra-Tight Infill Development”, we examine how surrounding development influences well performance, why traditional spacing metrics can fall short, and how better measures of interference can improve forecasting and development decisions.This podcast episode is based on the technical paper “Making an Impact: Machine-Learning-Ready Metrics for Ultra-Tight Infill Development”, authors: B. Davis, A. Qualls, K. Sathaye, G. A. Quintero. Download the paper here.

How does today's production impact tomorrow's inventory? In this episode of The Novi AI Roundup, we explore how prior depletion is reshaping future development opportunities in the Delaware Basin. Drawing from the technical paper “Assessing the Impact of Prior Depletion on Future Inventory in the Delaware Basin”, we examine how machine learning quantifies depletion effects on undeveloped locations, reveals changes in inventory quality, and helps operators make more informed development and capital allocation decisions.This podcast episode is based on the URTeC 2026 technical paper “Assessing the Impact of Prior Depletion on Future Inventory in the Delaware Basin”, authors: K. Sathaye, A. Toure, and D. Niederhut. Download the full paper here.

How does the impact of completion design change as a well matures? In this episode of The Novi AI Roundup, we explore how a multi-target machine learning model and SHAP values reveal the changing influence of completion parameters over a well's life in the Williston Basin. Drawing from the technical paper “Deriving Time-Dependent Scaling Factors for Completions Parameters in the Williston Basin using a Multi-Target Machine Learning Model and SHAP Values”, we examine how time-dependent scaling improves our understanding of production drivers, helping operators optimize completion strategies and build more accurate production forecasts.This podcast episode is based on the technical paper “Deriving Time-Dependent Scaling Factors for Completions Parameters in the Williston Basin using a Multi-Target Machine Learning Model and SHAP Values”, authors: D. Niederhut, K. Crifasi, K. Darnell, K. Sathaye, T. Cross. Download the paper here.

How can machine learning reveal the quality of the rock beneath the surface? In this episode of The Novi AI Roundup, we explore how GeoSHAP transforms machine learning outputs into a practical Rock Quality Index for unconventional reservoirs. Drawing from the technical paper “GeoSHAP: A Novel Method of Deriving Rock Quality Index from Machine Learning Models and Principal Components Analysis”, we examine how explainable AI helps identify the geological drivers of well performance, separate rock quality from operational effects, and improve development planning.This podcast episode is based on the technical paper “GeoSHAP: A Novel Method of Deriving Rock Quality Index from Machine Learning Models and Principal Components Analysis”, authors: D. Niederhut, T. Cross, J. Chaplin, K. Sathaye, A. Cui. Download the full paper here.

Can you accurately forecast water production before a well is drilled? In this episode of The Novi AI Roundup, we explore how machine learning is improving water production forecasting in the Williston Basin. Drawing from the technical paper “Predicting Water Production in the Williston Basin using a Machine Learning Model”, we examine how ML identifies the geological and operational drivers of water production, helping operators better plan infrastructure, manage disposal costs, and optimize field development.

This podcast episode is based on the technical paper “Predicting Water Production in the Williston Basin using a Machine Learning Model”, authors: D. Niederhut, K. Crifasi, K. Darnell, K. Sathaye, T. Cross. Download the full paper here.

How does spacing impact well performance over time? In this episode of The Novi AI Roundup, we explore how machine learning helps quantify the long-term effects of interwell spacing in unconventional reservoirs. Drawing from the technical paper “The Impact of Interwell Spacing Over Time: A Machine Learning Approach”, we examine how spacing influences production decline, recovery, and well interference throughout a well’s life, and why short-term production gains can sometimes come at the expense of long-term value.

This podcast episode is based on the technical paper “The Impact of Interwell Spacing Over Time: A Machine Learning Approach”, authors: D. Niederhut, J. Ramey, J. Reed, K. Darnell, K. Sathaye, T. Cross. Download the full paper here.

How do you separate operator skill from rock quality? In this episode of The Novi AI Roundup, we explore how predictive machine learning models are being used to benchmark operator performance in the Williston Basin. Drawing from the technical paper “Benchmarking Operator Performance in the Williston Basin using a Predictive Machine Learning Model”, we examine how ML normalizes for geology, spacing, and development conditions to uncover the true drivers of well performance, and why apples-to-apples benchmarking matters in mature unconventional plays.This podcast episode is based on the technical paper “Benchmarking Operator Performance in the Williston Basin using a Predictive Machine Learning Model”, authors: D. Niederhut, K. Crifasi, K. Darnell, K. Sathaye, T. Cross. Download the full paper here.