Industry
Quick links:
Powering the factories of the future with AI
The future of industrial performance is digital. If you run an industrial company, you simply cannot afford not to invest in Factory of the Future or Industry 4.0 technologies.
Probayes is helping industrial companies of all kinds navigate digital transformation by developing tools based on artificial intelligence.
Probayes develops tools that respond to industrial companies’ main challenges:
- Improving product quality by identifying and analyzing the factors influencing non-quality
- Improving industrial performanceby detecting process drift before it occurs
- Optimizing processes by recommending the most efficient equipment operating parameters
Algorithms tailored to the manufacturing and other industries
Probayes has been developing, testing, and implementing a wide range of algorithmic tools for use cases in the manufacturing and other industries for fifteen years. Our core approaches include:
Time-series analysis
To make the best possible use of time-stamped data, widely used in industrial scenarios
Model explainability
And explainable predictions, a fundamental aspect of the many use cases that require an understanding of the factors that lead to an abnormal or a desired situation
Anomaly detection
Particularly relevant for predictive maintenance, where there is not generally a substantial enough history of faults to teach supervised learning models
Automatic relearning
Often necessary, given that processes and equipment change over time
Unsupervised classification
Can be used to identify unknown behaviors and situations to solve problems in industrial use cases characterized by complex processes
Better results, together
We work hand in hand with your business experts
Industrial processes are particularly complex. Generally speaking, industrial projects are characterized by a large number of data sources. At Probayes, we work hand in hand with our customers’ experts to understand the many moving parts of the challenge at hand so that we can develop the best possible solution. We make communication a priority to ensure that we learn from each other throughout the project. Our customers come away with a better understanding of AI implementation in industrial scenarios, and our people acquire a deep understanding of the issues our customers face. Each project strengthens our long-lasting customer relationships.
Better results, together
thanks to our partners
Success stories
The EU MONSOON project
Bringing AI to the process industries: Rio Tinto-Aluminium Dunkerque & GLNPlast
The EU MONSOON project
Eramet
EasymAInt predictive maintenance solution
With EasymAInt, makes predictive maintenance simple. Just install plug-and-play sensors to monitor your data and equipment and you are ready to go!
The EU MONSOON project
Success story
Bringing AI to the process industries: Rio Tinto-Aluminium Dunkerque & GLNPlast
Challenge
Aluminium Dunkerque, a Rio Tinto company, manufactures anodes, which are one of the main components used in aluminum electrolysis. The performance of the electrolysis process depends on the den sity of the anode. Aluminium Dunkerque turned to Probayes for a process optimization solution to increase anode densities.
The project
Working closely with the customer’s experts, we tackled the project from two angles. First, we took around 40 continuously-monitored process parameters to build a classifier to identify times when the anode densities were low. Second, we integrated an explainability module into the classifier that generates, in real time, a list of the top five parameters that caused the lower anode densities.
We also built a recommendation engine powered by a regressor. Here is how it works: If it is determined that the anodes coming off the production line are low-density, the engine generates three recommendations to increase density. Finally, we developed a user interface re-training module so that the solution can adapt to the constantly-evolving process.
Results
The final version of the solution went live at Aluminium Dunkerque in the summer of 2019. It is expected to significantly reduce the number of low-density anodes produced, for estimated annual savings in the hundreds of thousands of euros.
The research that led to the development of the solution was presented at TMS annual meeting and a book chapter was published.
The EU MONSOON project to increase the efficiency of data-intensive processes is strategic for partners Rio Tinto and LIBERTY.
“Probayes, which is also a member of the project consortium, worked closely with process experts and operators to understand the target processes and associated data. Our data scientists were able to come up with process optimizations that were previously unknown, or, if they were known, they could not be implemented in a controlled manner. Probayes’ solid understanding of the use case and training of smelter employees contributed to the successful implementation of the project.”
Vincent Maigron
MONSOON Project Manager, Rio Tinto
Register for a webinar or watch the recordings
Improve your industrial performance with Artificial Intelligence
The future of industrial performance is digital. If you run an industrial company, you simply cannot afford not to invest in Factory of the Future or Industry 4.0 technologies.
AI in the loop: Optimize an industrial process
Listen to what LYNRED has to say about Probayes’ artificial intelligence tool for decision assistance in an industrial process use case.
Our customers*
*We take confidentiality seriously. Some of our customers’ names and other identifying information have been removed.
Copyright 2019 – Probayes – 53 avenue Kuntzmann – 38330 Montbonnot – France – Tel: +334 76 42 64