Real-time estimation of rock geomechanical properties plays a vital role in evaluating wellbore stability, optimizing drilling operations, completion design and maintaining formation integrity, thereby enhancing overall drilling efficiency. Conventional empirical correlations often lack the accuracy and real-time predictive capability required for key properties such as young's modulus. Furthermore, these correlations rely on input parameters such as compressional and shear wave travel times and bulk density that are typically measured approximately 90 ft behind the drill bit during drilling operations. This study presents a robust model for real-time prediction of a computed static Young's modulus while drilling, using near-bit drilling parameters and gamma ray measurements.
A total of 6,279 datasets from five wells in the Volve Field were collected for this study. The database consists of logging-while-drilling measurements and drilling parameters. The model predictors include standpipe pressure, rate of penetration, weight on bit, surface torque, gamma ray, true formation resistivity and equivalent circulating density. The input features were selected based on their proximity to the drill bit and their strong influence on the target variables. As an initial step, exploratory and preprocessing data analysis was performed to investigate the relationships between the input features and the target variables, and to remove outliers and abnormal data. Data from five wells were used for model development (randomly split into 80% for training and 20% for testing), while one additional well was reserved as blind validation sets. For model development, both supervised regression machine learning algorithms (Extreme Gradient Boosting and Random Forest) and deep learning algorithms (multilayer perceptron neural networks) were trained and optimized for real-time prediction of young's modulus.
The results show that the XGB, ET and RF models achieved high accuracy (R2 = 0.99 – 1.00) on the training dataset but experienced a notable decrease to approximately 0.91 on the validation dataset. The multilayer perceptron (MLP) achieved a consistent accuracy of 0.87 on both the training and validation datasets. Comparative analysis revealed that XGB, ET and RF exhibited mild overfitting, indicated by approximately 10% accuracy drop between training and validation. On the blind test, the optimized MLP model demonstrated superior generalization performance compared to RF and XGB. Model evaluation confirmed that the MLP accurately captured depth-dependent variations of young's modulus in offset wells, with maximum mean absolute percentage errors (MAPE) of 4.32%.
This study presents a robust and reliable machine learning framework for real-time prediction of young's modulus, reducing dependence on core testing and enabling dynamic decision-making during drilling operations. The proposed method offers a scalable approach for intelligent formation characterization and can be seamlessly integrated into real-time drilling advisory systems. Its implementation supports confident drilling and mitigates uncertainties related to wellbore instability, ultimately reducing non-productive time and overall operational costs.
R. Elgaddafi, B. M. Bitayib, K. Amadi et al.· SPE Nigeria Annual Internati...· 0 citations
Reliable forecasting of drilling rate of penetration (ROP) remains a technical challenge due to its complex dependence on several operational, geological, and directional conditions that are highly nonlinear and well-specific. These difficulties are amplified in deviated wells, where traditional empirical relations often fail to capture the combined effects of lithological variability and directional changes. This study develops a machine learning (ML) model for ROP prediction using multi-well field data and evaluates candidate ML algorithms/models not only on numerical accuracy but also on their ability to reproduce reasonable trend responses to changes in key drilling parameters. The objective is to establish a validation approach that integrates statistical performance, cross-well generalization, and sensitivity behavior consistent with drilling mechanics.
Field data from 18 wells were used to develop the model, while two additional wells were reserved for independent validation. A total of 91,140 drilling datasets, comprising 24 parameters/features, were collected. After preprocessing and selective outlier screening, 85,695 records were retained for model training and testing. Feature selection resulted in nine highly influential parameters: True Vertical Depth (TVD), Weight on Bit (WOB), rotational speed (Rotation), mud flow rates (FLOW), mud density, shale content (Shale), Sandstone content (Sandstone), inclination (Inc), and dogleg severity (DLS). Eleven supervised ML algorithms were evaluated, representing instance-based, tree-based, boosting-based, neural-network, and stacking-ensemble model families. Model performance was assessed using feature-importance ratings and statistical metrics, including train/test R2, MSE, RMSE, and MAE, which were used to select the best-performing model. Finally, sensitivity analysis was conducted to evaluate the selected model's ability to reproduce physically meaningful ROP responses.
Model comparison across the training and testing splits demonstrated strong overall predictive performance, with training R2 values ranging from 0.91 to 0.99 and testing R2 values ranging from 0.88 to 0.94. Among the evaluated models, the Gradient Boosting (GB) model provided the best overall performance, with training R2 = 0.9940, testing R2 = 0.9399, RMSE = 36.88, and MAE = 21.52. The GB model also showed strong agreement with the averaged feature-importance ranking and reproduced physically meaningful ROP sensitivity trends for the dominantly influential drilling parameters. SHAP and feature-importance analyses confirmed that TVD, FLOW, WOB, shale content, and rotation were the most influential variables controlling ROP. External validation on two unseen wells further demonstrated strong generalization, with R2 = 0.9694 for Well-19 and R2 = 0.96 for Well-20.
This study presents a physically informed framework for evaluating and selecting data-driven ML models for deviated wells. The approach combines conventional accuracy metrics with feature-importance consistency, reasonable trend prediction in sensitivity analysis, and strong agreement between model predictions and unseen-well data. Through this integrated evaluation, a model is identified that not only reproduces historical data accurately but also yields correct responses under a diverse range of varying input parameters. The proposed methodology establishes a reproducible basis for developing more reliable ROP forecasting tools for complex well trajectories.
Nayem Ahmed, Ramadan Ahmed, V. Soriano et al.· SPE/IADC Asia Pacific Drilli...· 0 citations
This study investigates an artificial intelligence (AI) based approach for real-time prediction of Young's modulus and UCS using drilling and logging data and reveals that neutron porosity, formation bulk density and Gamma Ray are the three most influential predictors.
K. Amadi, R. Elgaddafi, B. M. Bitayib et al.· SPE Nigeria Annual Internati...· 0 citations