Sanding Risk Screening Tool

A practical production-engineering workflow for estimating drawdown-based sanding risk using reservoir pressure, productivity index, and rock strength indicators. Inspired by Statfjord field studies, SPE-188008-MS, and SPE-58721-MS.

What this is (and isn't)

This is a screening-level tool for flagging wells that may be operating in a sanding-prone regime, using data engineers commonly have on hand: reservoir pressure, rate, productivity index (PI), and an estimated UCS.

It is NOT a substitute for:

• Mechanical Earth Models (MEM)
• Modified Lade Criterion or Mohr-Coulomb geomechanical analysis
• Finite Element geomechanical simulations
• Commercial sand-prediction software

Results should be read as "Screening-Level Sanding Risk Indicators", not definitive sand production predictions.

Concept

Field engineers often know reservoir pressure, rate, and PI more reliably than they know flowing bottomhole pressure (FBHP). So instead of requiring FBHP directly:

FBHP     = Pr - (Q / PI)
Drawdown = Pr - FBHP  =  Q / PI
SRI      = Drawdown / UCS          (Sanding Risk Index)

An optional Statfjord-style depletion-adjusted Critical Drawdown Pressure is also estimated:

CDP_est       = UCS * K - m * (Pri - Pr)
Safety Margin = CDP_est - Drawdown

Where K (UCS calibration factor, default 0.5) and m (depletion sensitivity factor, default 0.2) are user-adjustable. If a well's initial reservoir pressure Pri is not supplied, it defaults to the current Pr, which zeroes out the depletion term.

Risk classification (on SRI)

Category	SRI range
Green	SRI < 0.25
Yellow	0.25 ≤ SRI < 0.40
Orange	0.40 ≤ SRI < 0.60
Red	SRI ≥ 0.60

Input format

CSV with these required columns:

Column	Description
Well Name	Well identifier
Pr	Reservoir pressure (psi)
Q	Production rate (stb/day or equivalent)
PI	Productivity index (stb/day/psi)
UCS	Unconfined compressive strength (psi)

Optional columns:

Column	Description
Pri	Initial reservoir pressure. Defaults to Pr if omitted.
m	Per-well depletion sensitivity factor. Defaults to CLI --m (0.2).
K	Per-well UCS calibration factor. Defaults to CLI --k (0.5).

See sample_wells.csv for a worked example, including a well with an invalid PI (0) to demonstrate the tool's validation/flagging behavior rather than crashing.

Usage

pip install pandas numpy matplotlib

python run_screening.py --input sample_wells.csv --outdir output

# Override the default depletion/calibration factors globally:
python run_screening.py --input my_wells.csv --outdir output --m 0.25 --k 0.55

Outputs (written to --outdir)

• sanding_risk_results.csv — full summary + ranking table (Pr, Q, PI, FBHP, Drawdown, UCS, SRI, Risk Category, CDP_est, Safety Margin, Rank, and validation notes for any flagged rows), sorted highest risk first.
• drawdown_vs_ucs_scatter.png — Drawdown (x) vs UCS (y), colored by risk category.
• sri_bar_chart.png — SRI by well, colored by risk category, with the Green/Yellow/Orange/Red band thresholds marked as reference lines.

Console output also prints a ranked list and flags any invalid/implausible rows (e.g. PI ≤ 0, UCS ≤ 0, or FBHP < 0 implying drawdown exceeds reservoir pressure).

Validation example

The sample dataset includes this hand-checkable case:

Pr = 2000 psi, Q = 500 bbl/day, PI = 5 bbl/day/psi, UCS = 5000 psi

Expected:

Drawdown = 500 / 5        = 100 psi
FBHP     = 2000 - 100     = 1900 psi
SRI      = 100 / 5000     = 0.02   -> Green
CDP_est  = 5000 * 0.5     = 2500 psi   (Pri defaults to Pr, so depletion term = 0)
Safety Margin = 2500 - 100 = 2400 psi

Running python run_screening.py --input sample_wells.csv --outdir output reproduces these exact values in sanding_risk_results.csv.

Technical basis and references

• Hettema, M.H., Andrews, J.S., Blaasmo, M., and Papamichos, E. Sand Production in the Statfjord Field. JPT, 2006. Empirical relation: Pcd + m(Pri - Pr) = C; depletion's role in sanding behavior; distinction between rock failure, grain mobilization, and grain transport.

• Moreno, F.G., Guizada, P., Abdul Aziz, A.A., and Khanferi, R.N. Application of Critical Drawdown Pressure Prediction in Completion Design to Minimize Sanding in a Clastic Gas Reservoir in Saudi Arabia. SPE-188008-MS.

• Ong, S., Ramos, R., and Zheng, Z. Sand Production Prediction in High Rate, Perforated and Open-hole Gas Wells. SPE-58721-MS. CDP decreases as reservoir pressure declines; drawdown-to-strength ratio as a screening heuristic.

Disclaimer

This tool uses commonly available field data (reservoir pressure, rate, PI, estimated UCS) to produce a screening-level heuristic, not a rigorous geomechanical prediction. Results should be used to prioritize wells for further, more rigorous sanding evaluation — not as a stand-alone basis for completion or operating decisions.