This paper describes scenario generators and portfolio models for oil and gas exploration and production (E&P) planning. The problem is how to allocate funds to a set of oil recovery projects to achieve a desired tradeoff between risk and return. A mix of development and exploration projects is assumed, further stratified by reservoir depth and uncertainty level of reserves. The scenario generator incorporates distributions for seven uncertain reservoir parameters, obtained from engineering and geologic data in the Tertiary Oil Recovery Information System (TORIS) database ( Capital and operating expenditure data comes from the IHS Energy Discovery Module ( A slightly compressible-liquid tank model is used to model the yearly oil production from each well in each project. Oil price trajectories are generated using a mean reverting model. A complex income and tax calculation evaluates the net present value (NPV) of free cash flow, and a Monte Carlo simulation generates NPV scenarios for each project.

We describe and compare optimization models that minimize several risk measures, including variance, semivariance, expected loss, and loss probability, as well as models that maximize expected utility. Risk measure values, optimal project weights (assuming any level of fractional participation is allowed in all projects), and portfolio NPV distributions arising from the various risk measures are compared, and contrasted with results of widely used ranking procedures that ignore risk.

You can access this article if you purchase or spend a download.