Abstract We conducted a practical case study that aims at estimating the shapes and parameters of probability distributions for key cost, time and activity performance inputs required by a risk, resource, and value simulator to conduct a stochastic valuation of a new exploration asset.
We analyzed a sample of 73 shallow offshore fields in Australia retrieved from a global field-by-field database that includes reserves, production profiles, financials, valuation, breakeven prices, ownership and other key metrics for global oil and gas fields, discoveries and exploration licenses.
The reviewed facilities concepts includes 40 steel platforms, 2 concrete gravity-based developments, 10 projects with extended reach drilling, 4 FPSO, and 17 subsea tie-backs. The aggregate CAPEX of projects in the sample over 1965–2015 is USD 99.1 billion (in nominal terms), which is commensurate with the total asset size of the Australian offshore petroleum industry.
We estimate probability distributions for all full-cycle parameters required to generate Monte Carlo CAPEX, OPEX, and production profiles. In particular, these include facility CAPEX per unit peak, development phase duration and scheduling rules, drilling expenditures per barrel of oil equivalent, cost of exploration and appraisal wells, OPEX-to-CAPEX ratio, abandonment cost ratio, fraction of hydrocarbons produced yearly at plateau, fraction of hydrocarbons remaining at end-of-plateau, terminal production rate, and fraction pre-drilled wells. Most of the above were found to be log-normally distributed.
The paper presents a simple yet robust workflow relying on real industry data that enabled assessing value, risks, and uncertainties of upstream assets at exploration and early appraisal stages. It can be seamlessly extended to other basins owing to a rich coverage of the online field analogs database.
The ultimate outcome of the study—a probabilistic value assessment of new exploration asset—represents a modern appraisal methodology relying on a corporation-wide software platform.