Kohli , Abhinandan
An analytical and meticulous Petroleum Engineer with 10 years’ experience in Petrophysics, Geology and Development Planning. Experienced in Data Science/ML for data processing and predictive analytics. I am experienced in data visualization in Power BI, Spotfire and Tableau. In my current role, I apply machine learning algorithms to analyze geological data optimizing the recovery of reserves resulting in current and future savings of 3 million USD a year.
Seeking to leverage my deep knowledge of subsurface clubbed with big data modeling skills to help increase revenue and optimize returns in your esteemed organization.
Shell US | Sep 20 – to date
• Built a program for interpretation of log data and facies prediction using the 2D and 3D indexed and probabilized self-organizing map.
• Delivered AI-based solution for predictive analysis and volume forecasting for anomaly detection and predictive maintenance.
• a-SME for PP-Integrated Reservoir Modelling.
• Project Lead for abandonment and new energies project with a total expenditure of € 3.5 mln.
• Coaching the graduates and subsurface staff on Petrophysics, Geology and WRFM areas.
DATA SCIENCE OPPORTUNITY LEAD
Shell Netherlands | Jan 19 – Aug 20
• Lead the interpretation of well data and facies prediction for derivation of petrophysical parameters using The Kohonen algorithm.
• Built a program using hierarchical clustering with dendrogram outputs to predict the optimal number of facies.
• Lead multiple ML/AI initiatives across Shell business verticals
• Created a training program for Shell staff to use machine learning algorithms to influence adoption of AI/ML techniques to execute projects (using internal resources or external vendors) and deliver business growth.
Shell India, Netherlands | Mar’ 18 – Dec’ 18
• Deep dive of data analysis using PCA, LDA, Spectrum Analyser and MCA for many data types.
• Reconstruct missing or poor-quality measurements correcting for bad hole conditions, environmental effects, acquisition problems using multilayer perceptron technology KMOD.
• Reduce systematic error for the subsequent data pattern by using the learn-validate-apply model.
• Coach subsurface community on Multi-scenario data modelling.
LEAD DOMAIN PETROPHYSICIST
Shell Netherlands | Jun’ 16 – Feb’ 18
• Lead Domain Data Scientist for Fluid Model and Saturation Height Model implementation in Baker Hughes Jewel Suite’s software.
• Image data Interpretation using parameter prediction and log reconstruction using multilayer perceptron technology.
• Developed tool to make automated, data-driven decisions to improve planning and decision making with embedded economic runs.
• Created enterprise-wide software application via cloud computing to model subsurface data.
• Prepared reports, presentations, dashboards, and/or data visualizations for various audiences –up to and including the executive level.
Awards and Certifications
SPE Star Scholar Award (2010)
SPE Kazakhstan Best Paper (2011)
SPE Young Member Award (2015)
• 185387-MS SPE - Data Integration Helps in Identifying New Waterflood Opportunities in Western Desert
• 188594-MS SPE – Saturation Data Modelling Under Complex Fluid Fill History - Drainage and Imbibition
• SPE Poster - Optimizing Data Appraisal Strategies for Tidal Clastic Reservoirs in Western Desert, Egypt
• SPWLA 2020 - The Value of Data Integration in a Complex Well Scenario: A case Study from the West Netherlands
• Domain Expertise : Advanced PP workflows, Clastic and Carbonates Model Build, Open and Cased hole Log Analysis
• Python: NumPy, Pandas, Matplotlib, Seaborn, SciKit-Learn
• Machine Learning: Multi-linear regression, Decision trees & Random forest, Clustering techniques, Gradient boosting, Bayesian networks, Time series forecasting, PCA, LDA, KMOD
• Business intelligence: SQL, Tableau, Power BI, Excel
• Mathematical modelling and optimization: LP, MIP, Genetic Algorithms, Metaheuristics
• Other skills: R, NLP, Hypothesis Testing, Forecasting, Data Mining, Text Mining
• Software: Techlog, Rokdoc, Petrel, Jewel Suite, Plato
Indian School of Mines, Dhanbad, India (2007 – 11)
Project Management Program (Level D)
Project Management Institute
Python for Everybody Specialization
University of Michigan