Analysis of Decline Curve Models in Unconventional Wells and the Effectiveness of Modeling Early-time Data for the Purpose of Forecasting

There are currently several decline models in existence for production forecasting in unconventional reservoirs. This purpose of this research is to apply these different decline models to unconventional wells and provide insight on what models are most effective, particularly as they pertain to early-time data.

The findings of this study show that the Arps model with b equal to 2.0 and Duong model provide good fits that can be used for forecasting wells in linear transient flow. When early-time data is used to determine the parameters of the models, the fits of the curves are affected, but could still be potentially used as viable tools.

This study provides valuable data and clarity on how these decline models can be used or shouldn’t be used based upon the amount of data available. This clarification will be very valuable to reservoir engineers currently modeling unconventional reservoirs with only early-time data.