| name | production-forecaster |
| description | Forecast oil & gas well production using decline curve analysis. Use when estimating EUR, generating type curves, fitting Arps models (exponential, hyperbolic, harmonic), or running reserve calculations. Supports conventional and unconventional wells, P10/P50/P90 probabilistic outputs, and multi-field type curve comparison. |
| type | reference |
| version | 2.1.0 |
| category | data/energy |
| last_updated | "2026-03-08T00:00:00.000Z" |
| capabilities | [] |
| requires | [] |
| tags | [] |
Production Forecaster
Forecast oil and gas production using industry-standard decline curve analysis.
Supports Arps decline models, type curve generation, and EUR calculations for
reserve estimation.
When to Use
- Forecasting future production for oil and gas wells
- Estimating EUR (Estimated Ultimate Recovery)
- Generating type curves for field development planning
- Fitting decline parameters to historical production data
- Comparing well performance across different fields
- Supporting reserve booking and economic evaluation
Core Pattern
Historical Production → Decline Curve Fit → Parameter Estimation → Forecast → EUR
Decline Curve Models
Arps Equations
| Model | Equation | b value | Typical Use |
|---|
| Exponential | q(t) = qi * exp(-Di * t) | 0 | Mature conventional |
| Hyperbolic | q(t) = qi / (1 + b*Di*t)^(1/b) | 0–1 | Most common |
| Harmonic | q(t) = qi / (1 + Di*t) | 1 | Fracture-dominated |
| Modified Hyperbolic | Hyperbolic until D(t) ≤ D_min, then exponential | — | Long-term forecast |
Parameters: qi = initial rate, Di = initial decline rate, b = exponent, t = time.
Unconventional Models
Duong (transient linear flow): q(t) = q1 * t^(-m) * exp(-a/(1-m) * (t^(1-m) - 1))
Stretched Exponential: q(t) = qi * exp(-(t/τ)^n)
See references/examples.md for full class implementations and fitting code.
Key Classes
| Class | Purpose |
|---|
DeclineCurveAnalyzer | Fit decline models to production data; calculate EUR and validation metrics |
ProductionForecaster | Generate rate/cumulative forecasts from fitted parameters |
|
