Evaluation of Multi-Objective Methods for Portfolio Tradeoff Analysis

A multi-objective optimization algorithm was used to develop five-year capital plans for a portfolio of unconventional assets. The approach explored development plan tradeoffs, while considering statistical uncertainty of reserves additions. The objective was to apply the theory of multi-objective methods to an existing planning model, and to evaluate if such methods can provide greater knowledge, faster analysis, and better solutions.The tradeoff analysis workflow applied recent advances in constrained multi-objective optimization. A hybrid optimization algorithm, combining elements of mathematical programming and evolutionary computation, was used to calculate tradeoff frontiers under varying assumptions and constraints. Portfolio metrics included net present value, capital, reserve replacement ratio, capital effectiveness, production, reserves, and well counts. Statistical metrics were used to calculate uncertainty bounds (P10 and P90) on portfolio reserves additions, plus confidence factors for achieving given reserves targets. Constraints on drilling activity by area over time were also applied.The optimization algorithm successfully calculated tradeoff frontiers and efficient portfolios. The approach showed how tradeoff frontiers change with varying assumptions and constraints, and provided knowledge of strategic opportunities and limitations. The model enabled rapid replanning as objectives, constraints, and source data changed. The optimizer delivered better solutions than the previous planning model, and helped to identify short lists of optimized capital plans faster than otherwise possible. Reserves uncertainty metrics quantified the tradeoff between reserves targets and reserves confidence. The analysis revealed weaknesses of published uncertainty methods, and highlighted possible improvements for future research.

show abstract

Visualizing Portfolio Tradeoffs With Perceptually Accurate Self-Organizing Maps

Kostuik

El-Beltagy

et al. 2016

SPE Annual Technical Conference and Exhibition

View full text Add to dashboard Cite

Multi-objective optimization and unsupervised machine learning were used to visualize relationships between portfolio composition and business performance tradeoffs. A multiobjective global optimization algorithm reduced the exponentially large space of possible portfolio outcomes to a set of Pareto optimal tradeoff frontiers for various combinations of decision and outcome constraints. The high-dimensional decision space was then reduced with an unsupervised neural network to generate Kohonen self-organizing maps (SOMs) for each tradeoff frontier. The resulting decision SOMs were rendered with a perceptually accurate color map and interactively linked with the tradeoff frontiers to enable visual data mining. The interactive visualization revealed relationships between planning decisions and business outcomes in the complex and high-dimensional decision space. The interactive SOM automatically classifies similar strategies together and shows how neighboring strategies relate to each other. The analysis demonstrated that the proximity of portfolios on the tradeoff frontier does not imply similarity of underlying portfolio decisions, and that the mapping between decisions and outcomes becomes more complex with stricter constraints.

show abstract

Evaluation of Multi-Objective Methods for Portfolio Tradeoff Analysis

Cited by 3 publications

References 44 publications

Evolutionary Approaches for Project Portfolio Optimization: An Overview

Evolutionary Approaches for Project Portfolio Optimization: An Overview

Optimizing Unconventional Asset Development and Reserves Confidence With Portfolio Tradeoff Analysis

Visualizing Portfolio Tradeoffs With Perceptually Accurate Self-Organizing Maps

Contact Info

Product

Resources

About