The oil and gas industry uses production forecasts to make decisions, which can be as mundane as whether to change the choke setting on a well, or as significant as whether to develop a field. These forecasts yield cash flow predictions and value-and-decision metrics such as net present value and internal rate of return.

In this paper, probabilistic production forecasts made at the time of the development final investment decisions (FIDs) are compared with actual production after FIDs, to assess whether the forecasts are optimistic, overconfident, neither, or both.

Although biases in time-and-cost estimates in the exploration and production (E&P) industry are well documented, probabilistic production forecasts have yet to be the focus of a comprehensive, public study. The main obstacle is that production forecasts for E&P development projects are not publicly available, even though they have long been collected by the Norwegian Petroleum Directorate (NPD), a Norwegian government agency. The NPD’s guidelines specify that at the time of FID, the operators should report the forecasted annual mean and P10/90 percentiles for the projected life of the field.

We arranged to access the NPD database in order to statistically compare annual production forecasts given at the time of FID for 56 fields in the 1995 to 2017 period, with actual annual production from the same fields. This work constitutes the first public study of the quality of probabilistic production forecasts. The main conclusions are that production forecasts that are being used at the FID for E&P development projects are both optimistic and overconfident, leading to poor decisions.¹

¹The conclusions based on the analysis presented in this paper are limited to the set of fields from the NCS. However, other authors have demonstrated the optimism bias in production forecasts from fields around the world (Nandurdikar and Wallace 2011; Nandurdikar and Kirkham 2012).

石油和天然气行业使用产量预测来做出决策，这与是否改变油井的节流阀设置一样平凡，或者与是否开发油田一样重要。这些预测产生现金流量预测以及价值和决策指标，例如净现值和内部收益率。

在本文中，将开发最终投资决策（FID）时做出的概率生产预测与FID之后的实际生产进行比较，以评估预测是乐观的，过度自信的，还是两者都不是。

尽管在勘探与生产（E＆P）行业中在时间和成本估算上的偏差已得到充分证明，但概率性产量预测尚未成为全面，公开研究的重点。主要障碍是，即使很早就由挪威政府机构挪威石油局（NPD）收集了勘探与开发项目的产量预测，也无法公开获得。NPD的指南规定，在进行FID时，运营商应报告该油田预计寿命的预测年度平均值和P10 / 90％。

我们安排访问NPD数据库，以便统计比较FID时在1995年至2017年期间对56个油田的年度产量预测与该油田的实际年度产量。这项工作是对概率生产预测质量的首次公开研究。主要结论是，FID用于勘探开发项目的生产预测既乐观又过于自信，导致决策不力。^1个

¹基于本文提出的分析得出的结论仅限于NCS的领域。但是，其他作者也证明了世界各地产量预测中的乐观偏见（Nandurdikar和Wallace 2011； Nandurdikar和Kirkham 2012）。