The rates are not in dispute. The curve through them is. One of them fits an exponential and calls it conservative, one fits a hyperbolic with b pushed up near 1, and the third has a spreadsheet that has been copied through four jobs and nobody can say what it integrates. The gap between those answers is often larger than the volume anybody is arguing about, and none of the three can show their working line by line.
Paste the monthly history, give it an economic limit rate, and read the fitted parameters and the remaining volume off the same page as the arithmetic that produced them.
Least squares over the whole rate history returns qi, Di and b together, with the nominal and effective annual decline both stated. A deterministic sweep, so the same history always gives the same answer.
Exponential, hyperbolic and harmonic each have their own closed form, and the harmonic case is not the hyperbolic one with b set to 1 — that expression divides by zero. The tool switches form on the regime and tells you which it used.
Supply a terminal effective decline and the forecast switches to exponential at the month where the hyperbolic decline softens to it, summing the segment plus the tail. The switch month and rate are both shown.
If the history fits b ≥ 1 the integral diverges and there is no finite volume. The tool says so and names the reason, instead of quietly returning a number that depends on the horizon somebody happened to pick.
A 24-month Eagle Ford history, economic limit 250 bbl/month, cumulative to date taken as the sum of the history. Each regime is refitted to the same rates, so these are genuine alternatives rather than one fit re-integrated three ways. This is exactly the spread the argument is about.
| Model | qi bbl/mo | Di /mo | b | Remaining bbl | EUR bbl |
|---|---|---|---|---|---|
| Exponential | 12133.42 | 0.05902 | 0.000 | 45622.66 | 207285.66 |
| Hyperbolic (fitted) | 14499.96 | 0.12800 | 0.920 | 236792.02 | 398455.02 |
| Harmonic | 14945.42 | 0.14169 | 1.000 | 275198.12 | 436861.12 |
Every figure on the results page is printed with its formula and your own values substituted, so a partner, a bank or a counterparty can argue with your inputs rather than with a black box.
End rate, volume and running cumulative for every forecast year through to the economic limit, with the final cumulative reconciling to the remaining volume exactly.
History points and the fitted curve on a logarithmic rate axis, with the economic limit and the terminal switch marked, so a bad fit is visible rather than buried in a number.
Hold as many wells as you like, roll the EUR up across the field, and export the per-well summary plus every forecast row to CSV with the EUR carried through.
Installs to the home screen and runs with no signal. Your rates stay on your own device; nothing is uploaded anywhere.
Built for the reservoir and production engineer, the acquisition analyst and the working-interest partner who have to put a number on a well and then defend how they got it, by people who have worked the pads.
What it is: Arps empirical decline fitted to the rate history you enter and integrated to the economic limit you supply. It reports engineering estimates and shows every formula. It is not a material-balance, simulation or volumetric method, it does not assign any disclosure classification, and a fit on under 18–24 months of history is weakly constrained — the tool says so on screen rather than in the small print.