“Two Vital Secrets for Building Better Type Wells”

    Each year, companies use averaged well production (type wells) to support billion dollar expenditures to buy and develop oil and gas resources. These type wells often have unrepresentative rate-time profiles and recoveries over-stated by as much as 50%. These intolerable errors result from common, but incorrect, assumptions in constructing type well production profiles, and the selection and weighting of analog wells.

   Literature related to constructing type wells is sparse and incomplete. This lecture will fill that gap and lead participants to informed decisions for best practices in type well construction. Hind casting examples show that only small errors in recovery result when the type well construction combines historical and predicted production rates. This improvement results from using educated estimates (not intrinsic values) for months with no data to average, and from individual well forecast errors that offset one another. A Monte Carlo method incorporates risk and leads to better well selection and weighting factors, achieving more representative rate-time profiles. The recommended methodology incorporates aggregation and choosing different uncertain parameters. Parameter choice is important because it makes little sense to risk recovery (e.g., P90 for proved reserves) when the application demands a different parameter such as present value.

  Type well construction methods are common, but they have errors that are difficult to detect. Evaluators are likely using type wells for financial analysis, facility design, cash flow prediction, reserve estimation and debt financing without knowledge of the inaccuracies and options to improve accuracy.