PETROLEUM RISK AND DECISION ANALYSIS

Overview:
Good technical and business decisions are based on competent analysis of project
costs, benefits and risks. Participants learn the decision analysis process and foundation
concepts so they can actively participate in multi-discipline evaluation teams. The focus is
on designing and solving decision models. Probability distributions express professional
judgments about risks and uncertainties and are carried through the calculations.
Decision tree and influence diagrams provide clear communications and the basis for
valuing each alternative. Monte Carlo simulation is discussed and experienced in detail
in a hand-calculation exercise. Project modeling fundamentals and basic probability
concepts provide the foundation for the calculations. Mathematics is straightforward and
mostly involves only common algebra. Emphasis is on practical techniques for immediate
application. This is a fast-paced course and recommended for those with strong English
listening skills. This course is intended as the prerequisite for the Advanced Decision
Analysis with Portfolio and Project Modeling course.

Objectives:
How To:
• Describe the elements of the decision analysis process and the respective roles of
management and the analysis team
• Express and interpret judgments about risks and uncertainties as probability distributions and popular statistics.
• Represent discrete risk events in Venn diagrams, probability trees, and joint probability tables.
• Solve for expected values with decision trees, payoff tables, and Monte Carlo simulation (hand calculations).
• Craft and solve decision models.
• Evaluate investment and design alternatives with decision tree analysis.
• Develop and solve decision trees for value of information (VOI) problems.

Activities:
• The course covers the following topics:
• Decision Tree Analysis: decision models, value of information (a key problem type emphasized in the course), flexibility and control, project threats and opportunities
• Monte Carlo Simulation: Latin hypercube sampling, portfolio problems, optimization, advantages and limitations
• Decision Criteria and Policy: value measures, multiple objectives, HSE, capital constraint, risk aversion
• Modeling the Decision: influence diagrams, sensitivity analysis, modeling correlations
• Basic Probability and Statistics: four fundamental rules including Bayes’ rule, calibration and eliciting judgments, choosing distribution types, common misconceptions about probability
• Expected Value Concept: foundation for decision policy, features, pitfalls to avoid
• Implementing Decision Analysis: problem framing, guidelines for good analysis practice, team analyses, computer tools (discussion and demonstrations), mitigating risks
• Evaluating a multi-pay prospect (team exercise)