Open to all students of the IMT Schools from October 2 to December 22, this Data Science competition, organized by our School in partnership with Total, is part of an educational approach and is based on a problem encountered by the company’s business lines. The 2017 Challenge offers a challenge centered on “Well performance prediction.”
“Data Science as a whole relies on statistical and algorithmic methods from well-identified fields: statistical learning, machine learning, and data mining”
The objective of this challenge is to predict two performance indicators: the quantities of gas and condensate extracted from a well one year after its commissioning—that is, once the production system has stabilized.
All interested engineering students from our School are invited to register; the challenge continues until December 22.
Competitors must provide the best production forecasts for an oil well based on a dataset in order to optimize its operation.
From an educational perspective, the goal is to encourage the adoption of generic techniques in statistical learning and machine learning by fostering back-and-forth exchanges between theory and practice through this real-world case.
“Develop your models and find solutions”
Already 91 participants and 210 contributions; there are two months left to register and participate!
For a more complete description of the problem, types of files to be submitted, access to the rules, and other practical information, visit the challenge website here.
Winners will be selected based on both the predictive quality of their model and the evaluation of a report by a jury composed of faculty from IMT and representatives from Total. The top three students will receive an offer for a Data Science internship at Total.
Mines Saint-Étienne contact for the organizing team: Olivier Roustant, Fayol Institute
