Danial Arab, Ph.D. Email & Phone Number

Vancouver, British Columbia, Canada

North Vancouver, British Columbia, Canada

I developed both traditional ML models (Random Forest/XGBOOST classifiers fed with various CNN architectures as backbones) and deep learning models (U-net architecture) to quantify completion geometries and map perforation contours in three dimensions. This is achieved through semantic segmentation of ultrasound images of perforations at various frames.

Vancouver, British Columbia, Canada

I developed models to track the advancing interface in immiscible flow. I trained Random Forest/XGBOOST classifiers, fed with VGG16 as the backbone, to quantify displacement and classify wetting states. This approach allowed us to extract the meticulous details of the morphology and geometry of the advancing front.Dataset source: I performed microfluidic.

As an R&D Scientist, I was responsible for designing a workflow for the research projects, conducting HTHP experiments, analyzing the results, and presenting the project’s deliverables to the clients. I developed a detailed methodology to tune reservoir wettability through alteration techniques, which was recognized by industry experts and publications.

Ph.D. Dissertation Title: Understanding Immiscible Displacement Processes in Oil ReservoirsIn my Ph.D., I tried to understand immiscible displacement processes in the presence of surface-treated nanoparticles. I wrote a comprehensive critical review article, with more than 130 citations so far, which I am proud of. Also, my Ph.D. led to another 5 Journal.

Trondheim, Sør-Trøndelag, Norway

Microfluidic study of imbibition vs. drainage displacement

Tehran, Tehran Province, Iran

• Leveraged SQL to extract data from production logs, enabling efficient data retrieval, cleaning, and analysis
• Developed an automated data processing pipeline, reducing data preprocessing time by 50%
• Utilized statistical methods and machine learning algorithms to identify patterns, trends, and anomalies
• Implemented predictive models to forecast production issues and identify a critical production bottleneck to prevent potential downtime, leading to process optimization and a 15% increase in overall throughput.

I taught twice weekly Petroleum Production Engineering and Reservoir Engineering undergraduate courses.

I taught twice weekly Well Testing Analysis and Reservoir Engineering undergraduate courses.

Industrial Post-Doctorate, Computer Vision

Doctor Of Philosophy (Ph.D.), Petroleum Engineering, Gpa 4/4

Master’S Degree, Petroleum Engineering, First-Ranked In All Engineering Disciplines (Gpa: 19.41/20)