Abstract:Abstract The petroleum industry has always struggled with the "disconnect" between modeling subsurface uncertainties and integrating them into full robust economic analysis. Why? Because there is inconsistent use of probabilistic analysis within asset teams and functions, using disparate software tools, few of which can communicate with each other. Depending upon the project and the company, commercial analysts may use single cases, scenarios, decision trees, best estimates or risk weighted analysis combined into a composite case. Few, if any, have moved their project analysis to a fully integrated stochastic approach. Again, why? Because an "all singing all dancing" subsurface through commercial model does not exist. Different functions in an asset team use different software tools, few of which communicate with each other. In addition, there are few software applications which support fully stochastic analyses, which further complicates the task. As a result, many subsurface uncertainty characterizations get reduced to a few cases which are passed to the commercial analyst. Wouldn't it be best to integrate both workflows? This would allow the subsurface teams to preserve their characterizations while the commercial analysts interact with the reservoir distributions in their native states. Have the commercial analysts run their calculations using their models and their decision making tools so they can clearly see when, where and how decisions may change based upon all uncertainties. It is more than modeling with precision; it is about knowing when we should be changing our minds. This paper is designed to stir some thoughts in this area and showcase an example where stochastic analysis and decision trees are combined in three different ways within a single case to provide proof to the adage "the whole is greater than the sum of the parts". A live working example will demonstrate how correlations, full economic analysis and value of information can best be achieved by combining forces. The aim is to demonstrate the power of new ideas and workflows which can be implemented and used today in a practical way.

From Data to Stochastic Modeling and Decision Making: What Can We Do Better?

Improving Robust Decisions with Data

A survey of decision making and optimization under uncertainty

Decision-Making Under Uncertainty: Beyond Probabilities

Stochastic Nash Equilibrium Problems: Models, Analysis, and Algorithms

Robust Data-Driven Decisions Under Model Uncertainty

Stochastic Systems: Modeling, Optimization, and Applications

A data-driven approach for a class of stochastic dynamic optimization problems

Combining Data and Physical Models for Probabilistic Analysis: A Bayesian Augmented Space Learning Perspective

A Survey of Contextual Optimization Methods for Decision Making under Uncertainty

Improving non-deterministic uncertainty modelling in Industry 4.0 scheduling

Uncertainty in Modeling and Simulation

Geotechnical uncertainty, modeling, and decision making

Numerical Studies of Statistical Management Decisions in Conditions of Stochastic Chaos

Models and model value in stochastic programming

Stochastic heuristics for decisions under risk and uncertainty

Decisions, decisions, decisions in an uncertain environment

Integrating the Subsurface and the Commercial: A New Look at Monte Carlo and Decision Tree Analysis

Uncertainty quantification and optimal decisions

Optimization under uncertainty in the era of big data and deep learning: When machine learning meets mathematical programming

Data‐driven adaptive nested robust optimization: General modeling framework and efficient computational algorithm for decision making under uncertainty