Uncertainty Quantification for Petroleum Engineers Using Machine Learning – Transforming Subsurface Data into Confident Operational Decisions

By the end of the course, participants will be able to:

• Explain the main sources of uncertainty in petroleum-engineering data workflows and distinguish between deterministic prediction and uncertainty-aware prediction
• Apply Monte Carlo simulation to hydrocarbon volumetrics and interpret probabilistic outputs such as PDFs, CDFs, and P10/P50/P90 estimates
• Use conformal prediction to generate calibrated uncertainty estimates for regression, classification, and time-series forecasting problems
• Build machine-learning workflows for petroleum datasets including PVT, RTA, production, lithology, and flow-regime data
• Evaluate uncertainty-aware models using both point metrics and interval or set-based metrics such as coverage, mean width, Winkler score, and prediction-set size
• Compare classical statistical forecasting with modern machine-learning and deep-learning approaches for production time series
• Communicate uncertainty clearly through plots, tables, confidence intervals, and prediction sets that support better technical decisions

• Petroleum engineers involve in reservoir engineering, production engineering, drilling, completions, well intervention, and field development seeking to integrate machine learning and uncertainty-aware techniques into engineering workflows
• Reservoir engineers, reservoir modellers, geoscientists, geophysicists, and petrophysicists responsible for subsurface characterisation, reservoir evaluation, reserve estimation, and production forecasting
• Data scientists, data analysts, AI specialists, and digital transformation professionals supporting upstream oil and gas operations who wish to develop domain-specific expertise in petroleum engineering applications
• Production technologists and asset performance engineers involved in production optimisation, surveillance, and predictive analytics for mature and developing fields
• Multidisciplinary asset teams responsible for exploration, reservoir management, drilling, and production planning requiring more reliable and risk-informed decision-making
• Technical managers, team leaders, and decision-makers overseeing digitalisation initiatives, predictive modelling, and data-driven engineering strategies within upstream organisations
• Professionals and organisations transitioning from conventional deterministic modelling towards uncertainty-aware, risk-based, and decision-ready machine learning workflows to improve operational performance and investment decisions

Prerequisites:

Participants should ideally have a basic understanding of petroleum engineering concepts, including production data, PVT properties, reservoir variables, and well logs, as well as familiarity with fundamental machine learning concepts such as training and testing datasets, regression, classification, and model evaluation. Prior exposure to Python programming, Jupyter Notebook or Google Colab, and working with tabular datasets is recommended to maximise the learning experience. Participants should also have reliable internet access and a Google Drive account to support the hands-on exercises and collaborative activities throughout the course. An interest in advancing from conventional point prediction methods to robust, uncertainty-aware decision-making is beneficial. However, no advanced mathematical background in uncertainty quantification is required, as the course introduces all key concepts from first principles before progressing to practical case studies.

• Intermediate

The course is designed around a balanced blend of concise theoretical sessions and hands-on practical applications. Each module begins by explaining the underlying principles and rationale behind the uncertainty quantification methods before progressing to notebook-based demonstrations using realistic petroleum engineering datasets. Participants will learn how to interpret uncertainty outputs to support informed engineering decisions while developing reproducible workflows that can be readily adapted to their own field data. Every major topic is delivered as a complete end-to-end workflow, covering data preparation, exploratory data analysis, model development, uncertainty quantification, model evaluation, and the effective communication of results.

Your expert course leader is a Data Scientist based in Australia with a master’s degree in petroleum engineering and over 30 years of international experience spanning petroleum geoscience, reservoir evaluation, and data-driven decision support. His unique combination of extensive subsurface expertise and advanced data science capabilities enables him to bridge traditional petroleum engineering workflows with modern machine learning techniques, helping organisations transform complex data into actionable engineering insights. Throughout his career, he has held senior technical roles with leading energy companies including Baker Hughes, Brunei Shell Petroleum, KUFPEC Australia, Premier Oil Indonesia, Santos Ltd, Apache Energy, Enserch Exploration, Sun Oil Far East, Hall-Houston Malaysia, and Western Geophysical Company. At Baker Hughes, he led subsurface evaluations for the D18 Block offshore Sarawak, successfully identifying new field extension opportunities through integrated seismic interpretation, reservoir characterisation, and volumetric analysis.

At Brunei Shell Petroleum, he was responsible for production geophysics and reservoir surveillance for the Champion Field, one of Brunei’s largest producing assets. He led 4D seismic interpretation studies, supported waterflood optimisation, integrated multidisciplinary datasets for reservoir management, and developed well trajectories to improve production performance. His experience across exploration, field development, reserve estimation, and production optimisation has provided him with deep expertise in analysing large and complex petroleum datasets under conditions of geological and operational uncertainty. Building on this extensive petroleum background, he has transitioned into data science, applying machine learning, statistical modelling, and predictive analytics to solve engineering challenges across the upstream oil and gas value chain. His expertise includes data preparation, feature engineering, predictive modelling, uncertainty quantification, and the interpretation of model outputs for engineering decision-making. He combines practical industry experience with hands-on technical instruction, delivering engaging programmes that integrate realistic petroleum datasets, Python-based workflows, and real-world case studies.

Unlock the potential of your workforce with customized in-house training programs designed specifically for the energy sector. Our tailored, in-house courses not only enhance employee skills and engagement but also offer significant cost savings by eliminating travel expenses. Invest in your team’s success and achieve specific outcomes aligned with your organization’s goals through our expert training solutions. Request for further information regarding our on-site or in-house training opportunities.

In our ongoing commitment to sustainability and environmental responsibility, we will no longer providing hard copy training materials. Instead, all training content and resources will be delivered in digital format. Inspired by the oil and energy industry’s best practices, we are leveraging on digital technologies to reduce waste, lower our carbon emissions, ensuring our training content is always up-to-date and accessible. Click here to learn more.