Dragon Oil Annual AI-Hackathon

• Surface network production flow behaviour in the oil and gas industry is a complex system influenced by many factors, including fluid properties, pipeline configurations, and operational conditions. The theme for AI Hackathon II revolves around the application of AI to predict and optimise this behaviour, focusing on the following:

• AI Flow Prediction Model: Using historical data inputs (of four months of production), develop AI models that accurately forecast production flow (of subsequent two months) under varying events and activities in the field.
  • Total Flow Rate
  • Oil Flow Rate
  • Water Flow Rate
  • Gas Flow Rate
  • GOR
  • Pressure
  • Temperature
• AI identification of the reasoning for decline:
• Predict the potential reasons that caused the change in production in the last two months by learning from the first 4 months' data. Identify the strings that are suffering from these changes in production.
• AI Production Optimisation Model: Creating algorithms that can optimize flow rates (increasing), reduce bottlenecks, and maximize efficiency in surface network operation activities.

• The primary purpose of AI-Hackathon-II is to provide a platform for industry professionals, academic faculty and students, and AI enthusiasts to collaborate on complex challenges in the E&P sector, specifically focusing on:

• Innovation in AI Applications: Encouraging the development of novel AI-based methodologies and tools that can accurately predict and optimize surface network production flow behavior.
• Knowledge Sharing: Facilitating the exchange of ideas, techniques, and best practices among participants from various disciplines, including data science and engineers and scientists from all disciplines.
• Industry Advancement: Contributing to the E&P by delivering tangible, AI-driven solutions that can enhance operational efficiency, reduce costs, and improve safety in surface network operations.
• Networking and Collaboration: Building a strong network of AI professionals, researchers, and industry experts who can continue to collaborate beyond the hackathon to push the boundaries of AI in the energy sector.

• Anonymous flow data of one platform wells, manifolds, a separator, pumps and risers.

AI-Hackathon-II will be structured to maximise participant engagement and innovation output through the following phases:

Registration and Team Formation:

• Participants register as individuals or teams (maximum of 5 members per team).
• Cross-disciplinary team formation is encouraged to combine AI expertise with domain-specific knowledge. And they can be from multiple institutions and individuals.
• Create a team name that you will enter in the submission form, even if the team comprises one person. (Note: Try to make it a unique and with a special name).

Problem Statement Release:

• A detailed problem statement aligned with the theme will be released at the beginning of the hackathon.
• Participants will have access to relevant datasets to develop their solutions.

Workshops and Mentorship:

• A series of pre-hackathon workshops will be conducted to familiarise participants with the objectives and data.
• Mentors from academia and industry will be available throughout the event to guide teams and provide technical and strategic advice.

Development Phase:

• Participants will have three months to develop their solutions.
• Continuous support will be provided through dedicated communication channels, like FAQ, email, etc.

Submission and Evaluation:

• Solutions will be submitted as a working prototype or model, with a technical report, data, and a presentation. The Report has to be in the form of a Technical Paper, where all the team members must be listed as authors and co-authors and. Also, to honor the efforts of the AI-Hackathon Committee members in formulating the problem and guiding through the solutions in emails, workshops, FAQs, and then testing the models, they are considered co-authors.
• The AI Hackathon Committee, comprising AI experts, industry leaders, and academics, will evaluate the submissions based on the criteria.

Final Presentations and Awards:

• Selected teams will present their solutions in a final round during the GOTECH 2025 conference.
• Awards will be given to the top-performing teams.
• The prize is a Trophy and certificate, while only students from the winning teams will get other physical reward).

  • (Note: For the winning team, only the students will get the Ticket and Hotel paid by the GOTECH 2025 conference.)

• Mean Absolute Error (MAE): MAE is calculated as the average of the absolute errors between predicted and actual values. The error is the difference between the actual value and the expected value. Therefore, MAE would be considered the primary criterion for ranking the models. Also, Root Mean Square (RMS) would be considered as an additional criterion where we encourage participants to report it also.
• Model Learning Time: This is the time needed to train a model on the training data set. Model Run Time is the second criterion when two algorithms deliver the same MAE. Therefore, it might be used in the MAE tie between different participant models.
• The ability to solve all the problems of all the wells and the dependent parameters, or not.

AI Hackathon II is committed to fostering an inclusive and respectful environment. All participants are expected to adhere to the following principles:

• Integrity: Maintain honesty and transparency in all aspects of the competition.
• Respect: Treat all participants, mentors, and organizers respectfully and courteously.
• Collaboration: Encourage teamwork and knowledge sharing within and across teams.
• Professionalism: Demonstrate professionalism in conduct, submissions, and presentations.

Once a participant submits the application form and downloads the data, another email will be sent to the participant after a few weeks with a link where the participant should:

• Upload the AI Models. The models must be in a format that can run in a Python IDE or Python compiler; therefore, all the dependent libraries (requirements.txt file) to run the model must be provided.
• Upload the prediction work report (the report should contain the following sections [Title, Participants and Affiliation, Abstract, Introduction, Method, Results, Discussion, Conclusion].
• Upload the input/output data of blind data set prediction used in the prediction in an xls sheet. Make sure that you provide unscaled results.
• All the data, models, and reports should be included in one zip file following this file naming format:                                                                                                [“Team Name”-“MMM-DD”-GOTECH2025AIHackthon.zip].

- Announcement of Hackathon: November 2024

- Submission Deadline for Predictions: January 31, 2025

- Evaluation and Validation Period: February 2025

- Winner Announcement during the opening ceremony (Winners will be informed early-on to prepare for the travel, visa, etc.): GOTECH 2025 (Apr)

For any additional information or query, please contact:

Gehan Megahed gmegahed@spe.org

Omar Alfarisi oalfarisi@dragonoil.com

Muhammad Gibrata  mgibrata@dragonoil.com

Maryam Dashtebayaz mdashtebayaz@dragonoil.com

• Application Form: Please use this form to submit your application. Once you click submit, a new page containing the link to download the data will appear. https://forms.office.com/r/QRa7aL1kgN
• Upload Response: please use this link to upload your final response in one zip file as mentioned in the Additional Notes for Responding section.

https://sway.cloud.microsoft/oPGqkNjhD9znySyN?ref=Link&loc=play