Probabilistic Decline Curve Analysis

Fully automated probabilistic DCA leveraging state-of-the-art machine learning and data assimilation approaches for conventional and unconventional wells.

Sign up for a free trial below!

Estimate P10-P50-P90 decline curves accounting for data noise and uncertainty

Automatically remove outliers for more robust and reliable decline curve analysis

Use traditional Arp’s decline models or many more advanced models

Use waterflood specific (WOR vs Np) or unconventional specific models (Duong, etc.)

Automatically detect well events and use superposition to build more reliable models

Fully automated model fitting, hundreds of wells can be fit in minutes

Correct decline curve analysis for changing bottom hole pressures

Perform DCA at well level, layer level or on arbitrary aggregations

Can run locally or on Tachyus Cloud, with no infrastructure footprint

Data can be exported easily to other visualization and analysis tools

Tachyus’ Probabilistic Decline Curve Analysis leverages state-of-the-art machine learning and data assimilation approaches for faster, more accurate and reliable production and reserves forecasting.

How it works

Decline curve analysis (DCA) is one of the oldest methodologies used by production and reservoir engineers to predict future production, calculate reserves and estimate ultimate recovery, and it is fundamental to economic valuation of oil and gas assets. In practice, DCA is usually performed using deterministic tools that provide a unique solution ignoring the probabilistic nature of the production data due to measurement and back allocation errors.

Leveraging state-of-the-art machine learning and data assimilation approaches with traditional and advanced decline models , Tachyus’ Probabilistic Decline Curve Analysis (pDCA) creates an ensemble of models to automatically fit the historical data and estimate future production and reserves in a probabilistic manner providing P10-P50-P90 estimates of these entities. Additionally, Tachyus pDCA automatically identifies and removes outliers, improving the accuracy of the models, can also correct for BHP variations and results can be aggregated to arbitrary groups of wells.



Watch a brief example of how to add a new case in the app

Sign up now for a 30 day trial!