Press

Connecting refinery optimization silos with closed-loop AI

Gil Cohen's February 2021 feature in Hydrocarbon Processing on connecting a refinery's separate optimization silos with closed-loop AI to recover margin.

Refining
Read at hydrocarbonprocessing.com
Wireframe illustration of a crude unit, catalytic cracker and reformer linked by signal lines to one central node

Imubit CEO and co-founder Gil Cohen wrote this feature for the February 2021 digital transformation issue of Hydrocarbon Processing. Its opening point is that refineries and chemical processing plants are complicated to model, control and optimize. The feature argues that refinery optimization is often split into separate silos, and that connecting them with closed-loop AI recovers margin.

Hydrocarbon Processing asks readers to log in or subscribe to read the full article. For how Imubit applies Map, Model and Control across refinery units today, see refining.

Read the feature in Hydrocarbon Processing

Read at hydrocarbonprocessing.com

Related reading

All articles
Portrait of Gil Cohen, CEO and Co-founder of Imubit
Team

Gil Cohen

Gil Cohen, CEO and co-founder of Imubit, trained as a mathematician and control engineer and has led the company since its founding in 2016.

Wireframe illustration of a crude unit, catalytic cracker and steam cracker linked by a network of signal lines
Press

Executive Viewpoint: interconnected neural networks in plant optimization

Gil Cohen's Executive Viewpoint column in the March 2021 issue of Hydrocarbon Processing, on interconnected neural networks for plant optimization.

Wireframe illustration of a catalytic cracker, hydrocracker and reformer feeding a gasoline blend header and product tanks
Article

Four refinery operations issues that erode margin

Unit-to-unit variability, slow response to market swings, reactive maintenance and control room bandwidth limits: four ways refinery margin leaks.

Where to start

Find out what your gap to plan is costing you.

  • One to two hours for your team to prepare and upload the data package

  • Results within a few working days of approval

  • No workshops

Send your data as it is. Gaps, outages and noisy tags are normal, and our engineers screen and clean them in the platform. If anything limits a use case, we tell you.

Let’s get startedHow a pilot works