Fundamentals of AI & Data Analytics in Oil and Gas is a practical 3-day course that demystifies artificial intelligence, machine learning, and data analytics for professionals operating in the petroleum sector. No prior programming or data science background is required. The course explains how these technologies actually work, where they are being deployed across the oil and gas value chain, and — crucially — how regulators, government officials, and industry professionals can evaluate, govern, and leverage them effectively.

The course culminates in two hands-on workshop sessions covering the industry’s leading data visualisation and analytics tools: Tableau, Spotfire (facilitator demo), and Power BI on Day 2, and Pivot Tables with Power Query in Excel on Day 3 — equipping participants with immediately applicable skills.

 

Format	Live virtual sessions via Zoom or Google Meet
Duration	3 Days — 2 hours per session (1h45 delivery + 15 min Q&A)
Frequency	Hosted once a month
Certificate	Certificate of Completion awarded upon finishing all sessions
Open To	Petroleum professionals, regulators, analysts, government officials, energy enthusiasts
Contact	+1 365-654-9225 | info@pentoragroup.com | pentoragroup.com

 

Note: Days 2 and 3 include hands-on tool workshops. Participants should have Tableau Public (free), Power BI Desktop (free), and Microsoft Excel installed prior to Day 2. Spotfire will be demonstrated by the facilitator (10 min demo on Day 2 — no installation required).

 

Learning Objectives

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

• Explain the core concepts of Artificial Intelligence, Machine Learning, and Deep Learning without requiring a mathematical background.
• Identify and evaluate the main AI and data analytics applications across the oil and gas value chain: exploration, drilling, production, and HSE.
• Understand how regulators and government bodies can use data analytics to strengthen petroleum sector oversight and compliance monitoring.
• Critically assess AI models and data tools presented by petroleum operators — asking the right questions about data quality, bias, and model validity.
• Apply Tableau, Power BI, and Spotfire to visualise petroleum production and operational data.
• Build Pivot Tables and use Power Query in Excel to clean, transform, and analyse petroleum datasets.
• Formulate data governance principles for the responsible use of AI in petroleum regulation and operations.

 

Day 1 — AI & Machine Learning: Concepts and Applications in Oil & Gas

Session Overview

This foundational session introduces participants to artificial intelligence and machine learning without requiring any prior technical background. It traces the AI landscape from basic definitions to the most advanced large language models, and then maps these technologies systematically onto the oil and gas value chain — from seismic interpretation to refinery optimisation.

 

1.1  The AI Landscape: From Algorithms to Large Language Models

• What is Artificial Intelligence? A clear, jargon-free definition and the spectrum of AI capabilities
• Machine Learning (ML): how systems learn from data rather than being explicitly programmed
• Supervised learning: training a model on labelled examples — predicting well production, classifying reservoir lithology
• Unsupervised learning: finding hidden patterns — clustering fields by production behaviour, anomaly detection
• Deep Learning and Neural Networks: intuition without the mathematics — image recognition of seismic data
• Large Language Models (LLMs): what ChatGPT, Claude, and Gemini are — and what they can and cannot do for the petroleum industry
• The AI hype cycle: separating transformative applications from overpromised solutions

1.2  AI in Upstream: Exploration and Drilling

• Seismic interpretation augmented by AI: automated fault detection, horizon picking, and reservoir characterisation
• Drilling optimisation: real-time Rate of Penetration (ROP) optimisation using ML algorithms
• Early kick detection: using mud logging data and ML to detect well control events before they escalate
• Prospect ranking and portfolio optimisation: ML for prioritising exploration opportunities
• Digital rock physics: using AI to extract petrophysical properties from core images
• Limitations: what AI cannot replace — geological judgement, regulatory compliance, and ethical accountability

1.3  AI in Production and Reservoir Management

• Production optimisation: real-time choke and ESP (Electric Submersible Pump) optimisation using ML
• Decline curve analysis augmented by neural networks: more accurate production forecasting
• Physics-Informed Neural Networks (PINNs): combining reservoir simulation with machine learning
• Digital twins: creating virtual replicas of wells and facilities for scenario planning and optimisation
• Predictive maintenance: anticipating equipment failure before it happens — compressors, pumps, separators
• Integrity management: AI-powered detection of casing leaks, pipeline corrosion, and facility anomalies

1.4  AI for the Regulator: Surveillance, Compliance, and Governance

• Automated processing of operator reports: detecting anomalies and deviations from approved Field Development Plans (FDPs)
• Comparative benchmarking: using ML to identify underperforming fields relative to peer groups
• Satellite and remote sensing: AI-powered detection of flaring, spills, and unauthorised activity
• Natural Language Processing (NLP): extracting insights from thousands of technical documents automatically
• Case study: how NUPRC (Nigeria) and the OGA (UK) are using data analytics for regulatory oversight
• Building a data-driven regulatory function: what a modern petroleum regulator needs in terms of data, tools, and skills

 

AI in Oil & Gas: Scale of the Opportunity ▶  McKinsey estimates AI could generate $1–4 trillion in value annually across energy industries by 2030. ▶  BP has reported a 30% reduction in non-productive drilling time using AI-powered real-time decision support. ▶  Saudi Aramco’s GigaPower digital transformation uses AI across its entire value chain — from seismic to retail fuel. ▶  Africa’s challenge: data scarcity, legacy infrastructure, and skills gaps mean AI adoption requires a phased, pragmatic approach. ▶  The regulator’s opportunity: AI can dramatically improve oversight capacity without proportional increases in headcount.

 

Day 2 — Data Governance & Hands-On Workshop: Tableau, Spotfire (Demo) & Power BI

Session Overview

Day 2 opens with a focused session on data governance and the critical evaluation of AI models — essential skills for any petroleum professional or regulator engaging with operator-submitted analytics. The session then transitions into a practical workshop: a 10-minute facilitator demonstration of Spotfire, followed by 45-minute hands-on exercises in Tableau and Power BI, applying each tool to a petroleum production dataset. A 15-minute Q&A closes the session.

 

2.1  Data Quality and the Foundation of Reliable AI

• The ‘garbage in, garbage out’ principle: why data quality determines model quality
• Common data problems in petroleum: missing values, sensor errors, inconsistent units, and reporting gaps
• Data representativeness: when training data does not reflect operational reality — a dangerous AI failure mode
• Data ownership and access in petroleum regulation: who owns production data, and what should operators be required to submit?
• Building a data governance framework for a petroleum regulatory authority: key principles and practical steps

2.2  Algorithmic Bias and Model Explainability

• What is algorithmic bias? How training data encodes and amplifies historical inequities and errors
• Examples in petroleum: production models trained on North Sea data, applied uncritically to West African fields
• Explainability (XAI): why black-box models are dangerous in regulatory and high-stakes operational contexts
• SHAP and LIME: a conceptual introduction to the tools that make AI models interpretable
• When operators present AI models to regulators: a practical evaluation checklist
• Emerging AI regulation: the EU AI Act, OECD principles, and African frameworks — implications for petroleum governance

2.3  Workshop — Tableau (45 minutes, hands-on)

• Introduction to Tableau Public: interface, connecting to data, and building your first visualisation
• Hands-on exercise: importing a petroleum production dataset (provided) and creating a time-series production chart
• Building an interactive dashboard: filtering by field, operator, and date range
• Maps in Tableau: plotting well locations and production volumes on a geographic canvas
• Export and sharing: how to publish and share Tableau dashboards for regulatory reporting

2.4  Facilitator Demo — Spotfire (10 minutes)

• Introduction to TIBCO Spotfire: why it is the dominant analytics platform in upstream petroleum operations (no installation required — facilitator demonstration)
• Facilitator demonstration: navigating a production and well performance dataset in Spotfire
• Key Spotfire features: cross-filtering, data relationships, and the ‘marking’ interaction model
• Spotfire for the regulator: monitoring field-level production trends and comparing actuals against FDP forecasts
• Hands-on exercise: creating a production decline visualisation and identifying anomalous wells

2.5  Workshop — Power BI (45 minutes, hands-on)

• Introduction to Power BI Desktop: the Microsoft analytics platform and its integration with Excel and SharePoint
• Connecting to data sources: Excel files, CSV, and direct database connections
• Hands-on exercise: building a petroleum KPI dashboard — production volumes, uptime, and cost per barrel
• DAX fundamentals: writing simple calculated measures for cumulative production and year-on-year comparisons
• Publishing and collaboration: how Power BI reports can be shared across a government ministry or regulatory authority

 

Tool Selection Guide: Which Tool for Which Job? ▶  Tableau: Best for rich, interactive visual storytelling and public-facing dashboards. Easiest learning curve for non-technical users. ▶  Spotfire: Industry standard in upstream petroleum — purpose-built for well and production data. Preferred by IOCs and NOCs for operations analytics. ▶  Power BI: Best for organisations already in the Microsoft ecosystem (Excel, SharePoint, Teams). Most cost-effective for government bodies. ▶  Recommendation: For a petroleum regulator, Power BI + Spotfire (demo access) is the most pragmatic combination given licensing costs and data volumes.

 

Day 3 — Excel for Petroleum Data Analysis: Pivot Tables and Power Query

Session Overview

The final session is a focused 1h45-minute hands-on Excel workshop, with 15 minutes reserved for Q&A,, providing participants with immediately applicable skills for petroleum data analysis using tools already available in most government and corporate environments. The session covers Pivot Tables for production data summarisation and Power Query for data cleaning, transformation, and automation — without requiring any programming knowledge.

 

3.1  Why Excel Still Matters in Petroleum Analytics

• Excel’s continuing dominance: why the petroleum industry — including regulators, NOCs, and small operators — still relies heavily on spreadsheets
• The gap between Excel’s perceived and actual capabilities: most users access less than 20% of its analytical power
• Excel vs. Python vs. specialised tools: when Excel is the right choice and when to escalate to more powerful platforms
• Setting up your Excel environment for petroleum data analysis: recommended settings, add-ins, and data organisation principles

3.2  Workshop — Pivot Tables for Production Data (50 minutes, hands-on)

• What is a Pivot Table? The core concept and why it is the single most powerful Excel feature for data analysis
• Hands-on exercise: creating a Pivot Table from a monthly well production dataset (provided — 500+ rows)
• Summarising production by field, operator, and year: group, filter, and drill down
• Calculated fields: adding cost-per-barrel and cumulative production columns inside the Pivot Table
• Pivot Charts: turning your Pivot Table into a dynamic, filterable production visualisation
• Slicers and timelines: making your Pivot Table dashboard interactive without any coding
• Common Pivot Table mistakes: wrong aggregation, data refresh issues, and blank row handling

3.3  Workshop — Power Query for Data Cleaning and Transformation (50 minutes, hands-on)

• What is Power Query? Microsoft’s built-in ETL (Extract, Transform, Load) engine — no code required
• The petroleum data cleaning challenge: inconsistent date formats, merged cells, missing values, and multi-sheet operator reports
• Hands-on exercise: importing a messy operator monthly report and cleaning it with Power Query
• Key Power Query transformations: remove duplicates, split columns, fill down, unpivot, and merge tables
• Combining multiple monthly files automatically: how Power Query can consolidate 12 months of operator data in one click
• Refreshing your analysis: when new data arrives, your entire cleaned dataset and Pivot Table update with one click
• Connecting Power Query output to Pivot Tables: building an end-to-end automated reporting pipeline in Excel

3.4  Building a Regulatory Production Monitoring Dashboard in Excel

• Putting it all together: using Power Query + Pivot Tables + Pivot Charts to build a production monitoring dashboard
• Dashboard design principles: clarity, accessibility, and actionability for non-technical decision-makers
• Hands-on exercise: building a three-panel dashboard showing monthly production trends, field comparisons, and operator compliance status
• Adding conditional formatting: automatic red/amber/green status indicators for production vs. FDP targets
• Protecting and distributing the dashboard: locking structure while allowing authorised data refresh

3.5  Synthesis and Next Steps

• Connecting the tools: how Tableau, Power BI, Spotfire, and Excel fit together in a petroleum analytics ecosystem
• The data skills roadmap: from Excel proficiency to Python and R — a realistic learning path for petroleum professionals
• Building a data-literate team: recommendations for training, tool adoption, and change management in a regulatory authority
• Q&A with facilitator — open discussion on applying these tools to participants’ specific regulatory and operational contexts
• Course wrap-up, key takeaways, and certificate issuance

 

Why 2 Hours Per Session (1h45 + 15 min Q&A)? ▶  Spotfire has no free desktop version — it is presented as a 10-minute facilitator demonstration, freeing up full 45-minute slots for hands-on Tableau and Power BI exercises. ▶  On Day 3, Pivot Tables and Power Query each receive 50 minutes of hands-on time — enough to complete the end-to-end production monitoring dashboard exercise. ▶  2 hours per day across 3 days (6 hours total) — with 1h45 of structured delivery and 15 minutes of Q&A — gives each tool and concept the time it deserves. ▶  Participants who complete the pre-session software installation and review the provided datasets in advance will get the most out of each workshop.

 

Assessment & Certification

Participants are evaluated on attendance and engagement across all three sessions, including active participation in the two workshop days. A Certificate of Completion is awarded to participants who attend all three sessions and complete the hands-on exercises. The certificate is issued by Pentora Group and recognises foundational competence in AI, data analytics, and petroleum data tools.