The conference will take place October 19th - 20th at Thon Hotel Slottsparken, Oslo

This year’s main theme  is “Scaling AI from proof of concept to operations”, and this title indicates that we focus on practical, operational use of a potential high-flying technical opportunity which is on everybody’s lips nowadays.

NFEA is arranging an important meeting place where we will cover the development of AI applications with emphasis on use cases from our industries, and we are just as eager to learn about the less effective use-cases as we are to hear the success stories.

A lot of our large industrial players  will share their experiences, and will also inform us on their way forward – how will this technology bring their business further?

Join us for a two days eye-opening conference – you will not regret it!

The program is under development, but here is some of the lectures:

• Operationalization of ML models in Borregaard
  "Moving modeling projects from the drawing room to value-creating processes requires patience and expertise. Not all projects show value in operation. We explain how we carry out such processes. There will be examples of projects that have been successful, and those that were not."
• Deep Learning / AI - Velux case (SICK AS)
  "With Deep Learning, programmable sensors are enabled to automatically make decisions based on classified images. The machine can then itself make decisions that are otherwise difficult to automate with traditional rule-based tools for machine vision. We use a concrete case for the classification of wood to exemplify a Deep Learning solution."
  
• COGNITWIN - Cognitive plants through proactive self-learning hybrid digital twins (Cybernetica AS)
  "In the COGNITWIN research program we develop a platform to allow process industry to efficiently deploy digital twin-based technologies. The research is pilot-driven, meaning that the developed technologies have been proven in real-life environments. In this presentation, we will present some of the results from the COGNITWIN pilots."
  
• Cloud-based virtual flow metering system powered by a hybrid physics-data approach for water production monitoring. 
  "In gas fields, continuous water production monitoring per well is key to promptly identify wells experiencing water breakthrough, plan for future interventions, plan the use of inhibitors, evaluate the fluid composition, or adjust the water-handling capabilities in the processing facilities. A cloud-based virtual flow metering (VFM) system using a hybrid physics-data approach is implemented. This approach is based on the description of the flow through the wellbore using physics-based models where a data-driven algorithm is used to tune the flow model. This implementation accounts for changes in the well performance and increase in water production, resulting in a self-calibrating solution."
• Lessons Learned from Implementing Machine Learning in an Industrial Context – Case study of carbon anode core temperature measurements 
  "Idletechs will in this talk present lessons learned from implementing such as system in an industrial context. Our experience is that building the machine learning model itself is only a small part of creating a successful solution that will survive in an industrial setting and that will be used by the people it is designed for. Here are some questions we’ve had to ask ourselves as part of the journey from idea to operations: How do you determine how simple or complex a model should be? What is good enough? How should raw sensor data be interpreted? Can you use the data directly? Where and how do you install the solution? How should results be presented? Where? How do you design a solution that people WANT to use?"
  

PROGRAM COMMITTEE
Bernt Eldor | Kairos Technology AS (Chair)
David Anisi | NMBU
Bertil Helseth | Intelecy AS
Frank Rørtvedt | Siemens AS
Marianne Ytterbø | Yara Norge AS
Karin Sundsvik | NFEA
Tonje Olsen | NFEA

The program will be available H E R E  as soon as possible