Petroleum companies commonly use a "bottom up" approach for constructing and evaluating their petroleum portfolios, i.e. the "fill the data cube approach." Extensive information for each well is coded into the economic software including but not limited to rock and fluid properties, production profiles, operating and capital costs and then aggregated to fields, business units and finally the total petroleum asset base. This approach requires a significant commitment in technical, managerial and computer resources "to fill the data cube;" seriously limiting portfolio analysis to companies willing to commit to those resources.
An alternative to the "bottom up" approach is the "top down" approach. Financial fund managers on Wall Street use this "top down" approach for the construction of financial portfolios. The financial analysts are first involved in making forecasts for the economy, then for industries and finally for companies. This same concept and process can be applied to the petroleum industry, greatly reducing the effort to prepare for and conduct portfolio analysis.
One of the touted benefits of the "bottom up" approach is that it preserves the uncertainty in reservoir and economic properties and this uncertainty is included when conducting portfolio analysis. However, at some point in the analysis, one must realize all that is being manipulated are aggregated distributions of properties. The central limit theorem suggests that these aggregated properties all tend toward a normal/lognormal distribution. The "top down approach" recognizes and uses this insight in constructing portfolios.
This insight is combined with advanced production decline curves methods to generate a production profile consistent with theoretical reservoir response. The advanced production decline curves include oil and gas phases for both radial and fractured flow regimes, thus assuring the modeled responses are based on sound reservoir engineering principals using probabilistic reservoir properties. The production response when combined with probabilistic economic calculation creates an integrated stochastic reservoir and economic model. These models can be combined to create a replicating portfolio of a company's asset base. Portfolio analysis is performed on this replicating portfolio, examining the current asset performance, as well as acquisition and divestiture candidates. Using this methodology significantly simplifies the process, reduces the staff and computing resources required, dramatically speeding the process. An asset class is further de-aggregated into subsets and portfolio analysis conducted on an expanded description of a company's asset portfolio for additional detail as required.