Decline curve analysis (DCA) is one of the most widely used methods in petroleum engineering for
estimating future production and ultimate recovery from oil and gas wells. By fitting mathematical
models to historical production data, engineers can forecast how a well's output will change over
time and calculate the estimated ultimate recovery (EUR) — the total volume
of hydrocarbons a well is expected to produce over its economic life. DCA is essential for reserves
booking, asset valuation, field development planning, and investment decisions.
Decline Models Explained
The Arps decline equations, introduced by J.J. Arps in 1945, remain the foundation
of DCA. They describe three decline behaviors controlled by the b-factor:
- Exponential decline (b = 0): The production rate drops by a constant percentage
each time period. This is the simplest model and works well for mature conventional wells under
boundary-dominated flow, where the decline rate stays steady over time.
- Hyperbolic decline (0 < b < 1): The decline rate itself decreases over
time, meaning the well falls off quickly at first and then flattens. This is the most common model
for conventional wells and is characterized by its b-factor — a higher b means a slower
flattening of the curve.
- Harmonic decline (b = 1): A special case of hyperbolic decline where the
decline rate decreases proportionally with the production rate. Harmonic decline produces the most
optimistic EUR forecasts among the Arps models and is sometimes seen in gravity-drainage or
water-drive reservoirs.
- Duong model: Developed specifically for unconventional (shale/tight) reservoirs,
the Duong model accounts for the long transient flow periods typical of hydraulically fractured wells.
Unlike Arps models that assume boundary-dominated flow, Duong uses a power-law relationship between
rate and cumulative production, making it better suited for the first several years of shale well
production where Arps often overpredicts.
When to Use Each Model
For conventional wells that have been producing long enough to reach
boundary-dominated flow (typically 6–12 months or more), exponential or hyperbolic Arps
models are usually appropriate. For unconventional wells (shale oil, tight gas,
coalbed methane), Arps models can overestimate EUR because they were not designed for the extended
transient flow regime. In these cases, the Duong model or Stretched Exponential Production Decline
(SEPD) provides more realistic long-term forecasts. As a rule of thumb, always compare multiple
models and look at the goodness-of-fit metrics before committing to a forecast.
How EUR Is Calculated
EUR is calculated by integrating the decline curve from the current date forward until the well
reaches an economic limit — the minimum production rate at which revenue
covers operating expenses. This is then added to the cumulative production already achieved. The
economic limit depends on commodity prices, lease operating expenses, royalties, and taxes, which
is why EUR should always be paired with an economic analysis.
Common Mistakes in DCA
- Fitting too early: Using data from the initial flush production or cleanup
period, before the well has stabilized, can lead to overly optimistic forecasts.
- Ignoring shut-ins and workovers: Periods of zero production due to shut-ins,
mechanical failures, or workovers should be excluded or accounted for, not treated as production
data points.
- Wrong b-factor: Using b > 1 in Arps equations implies the well will
never reach an exponential decline, leading to physically unrealistic infinite EUR. Most
engineers cap b at 1.0 or switch to exponential terminal decline.
- Single-model reliance: Running only one decline model and assuming it is
correct. Always fit multiple models, compare R-squared values, and sanity-check against
analogous wells in the area.
All calculations in this tool run entirely in your browser — no production data is sent to
any server. Built by
Groundwork Analytics,
an AI and engineering company that builds digital tools and deploys AI agents for the energy industry. We help operators, service companies, and engineering teams automate workflows, optimize operations, and make better decisions with their data. Get in touch or email us at info@petropt.com.