Featured Project

Oil Production Wells Events Detection Using Change Points Detection Algorithms (Scale, Cease, Slugging Wells)

In the Rumaila supergiant oil field, detecting well issues like scaling, cessation, or slugging is critical to maintaining productivity. This project employs cutting-edge change point detection techniques to break down time-series data into sub-trends, each representing a distinct phase of well behavior. By analyzing statistical properties of these sub-trends, we identified problematic patterns, enabling swift action to maximize production uptime.

Featured Project

Petrophysical Data Imputation Using Advanced Statiscal And Machine Learning Algorithms

Missing petrophysical data can compromise reservoir characterization, leading to inaccurate decisions. This project developed state-of-the-art imputation methods tailored for oil and gas reservoirs, filling in the gaps with precision. The approach ensures reliable predictions of rock and fluid properties, empowering geoscientists with dependable data to guide exploration and development.

Featured Project

Unsupervised Production Anomaly Detection

Production anomalies can signal sensor errors or significant operational events. By applying advanced unsupervised techniques such as ADTK, Local Outlier Factor (LOF), and Isolation Forest, this project identified irregularities in Digital Oilfield (DOF) data. The result? Enhanced monitoring capabilities that highlight potential issues before they escalate, saving time and resources

Featured Project

Auto Machine Learning Platform For Subsurface Application.

Say goodbye to long coding sessions! This platform revolutionizes subsurface workflows by enabling engineers to build machine learning models without writing a single line of code. From data cleaning and feature engineering to anomaly detection and forecasting, every step is seamlessly executed through an intuitive drag-and-drop interface. Accelerate decision-making and focus on insights rather than scripts.

Featured Project

Oil And Gas HSE Virtual Advisor Using Large Language Model

Navigating complex HSE protocols in oil and gas can be daunting. This virtual advisor acts as a 24/7 assistant, delivering accurate, human-like responses to queries about safety guidelines, incident management, and best practices. Designed specifically for the industry, it ensures personnel can make informed decisions quickly, enhancing safety and compliance

Featured Project

Facies Classfication & Clustering

Understanding subsurface facies is fundamental to reservoir characterization. This project employed innovative clustering techniques to group similar geological facies, providing insights into reservoir heterogeneity. By automating facies classification, geoscientists can now interpret data faster and with greater confidence.

Featured Project

Permeability And Porosity Estimation

Permeability and porosity are the backbone of reservoir evaluation. This project introduced a robust machine learning framework to predict these properties with remarkable accuracy, transforming raw well data into actionable insights. The result is a deeper understanding of fluid flow and reservoir quality.

Featured Project

Production Forcasting Using Time Sereies Analysis

Forecasting oil production is critical for operational planning. Using advanced time-series analysis techniques, this project provided reliable production predictions, empowering operators to anticipate future trends and optimize field performance proactively.

Featured Project

Real Time Data Driven Framework for Rate of Penetration Optimization of S-Shaped Wells in a Southern Iraq Field Using Prior Knowledge

Drilling efficiency is vital in complex well trajectories. This framework leveraged real-time data and prior knowledge to optimize the rate of penetration in S-shaped wells. By combining data-driven insights with operational expertise, the project minimized downtime and reduced drilling costs in Southern Iraq fields.

Featured Project

Flow Regimes Characitimzation Using Supervied Machine Learning

Accurate flow regime identification is crucial for optimizing production. This project utilized supervised machine learning to classify flow regimes, providing operators with actionable insights to adjust production parameters and improve well performance.