Alfan Rizki Maulana

Undrained Reservoir Zone Identification in Mature Waterflooded Fields Using a Physics-Constrained Rule Score and Machine Learning

My thesis. How much of a waterflood's sweep can you reconstruct without running a reservoir simulator?

My bachelor thesis at ITB, supervised by Ivan Kurnia, Ph.D.

The answer is a hybrid rule score and machine learning framework. The deterministic rule score combines two physically interpretable features: a proxy for injector travel time, and a fault-aware Dijkstra distance to producer perforations. Both are mapped to hierarchical percentile rankings. Critically, every parameter is set from displacement physics rather than tuned against the answer key, which is the constraint that makes the result mean anything at all. The machine learning layer then predicts how those scores move as completions open, so the map updates on the day a perforation changes rather than on a calendar.

• Python
• resdata
• SciPy
• QGIS
• Volve
• Random Forest