Abstract
In the era of digital transformation, the petroleum industry faces a critical challenge: how
to extract actionable insights from vast amounts of data while optimizing operational
decisions. Nearly every aspect of our industry is, at its core, an optimization problem. Even
recent breakthroughs in Artificial Intelligence are largely driven by advances in
optimization algorithms.
One key challenge is balancing high-fidelity physics-based models with real-time data
assimilation. Traditional full-scale simulations are computationally expensive and
impractical for fast decision-making. Reduced Order Models (ROMs) provide a solution
by creating simplified representations that maintain accuracy while significantly reducing
computational costs.
We will explain how optimization algorithms in combination with data-driven reduced
order model can be used for decision-making, and how optimization could revolutionize
complicated well intervention workover scheduling using a portfolio management
approach.
Biography:
Justo Matheus, a Principal Engineer at SLB, has over 25 years of experience in
optimization, control systems, and machine learning for petroleum engineering. With a
Magister in Control Systems, he began his career in 1997 as an Automation Engineer and
later transitioned to become Principal Data Scientist.
He joined SLB in 2006 as a Senior Control Systems Engineer, he has been instrumental in
the development of Rotary Steerable Systems (RSS). For more than 15 years, he
specialized in Reduced Order Models (ROMs), Process Control, and Optimization; turning
complex physical phenomena into clear, data-driven models.
📍 Event Details
Location: TBD
Date: Tuesday, October 20, 2026
Time: 17:00
🤝 STAY AFTERWARDS
A networking reception with light refreshments will follow.
