Keeping kilns turning with machine learning
Published by David Bizley,
Editor
World Cement,
Mary Beth Kramer, CemAI, explains how artificial intelligence and machine learning are now essential tools for cement producers to stay competitive in an industry embracing process optimisation and automation.
Digitalisation is transforming the cement industry. Artificial Intelligence (AI)-based predictive maintenance is here and becoming essential for the competitive optimisation of manufacturing assets. AI is a driver in numerous industries, and it has proven to be a powerful tool for cement manufacturing, across the entire industry, independent of plant size, age, or equipment manufacturer. Enhancing the local plant’s maintenance functionality, AI provides increased benefits in terms of equipment reliability, availability, efficiency, and monitoring.
CemAI is an affiliate company of Titan Cement Group providing the cement industry with next-generation predictive maintenance solutions based on artificial intelligence. The solution is a unique mix of a proprietary licensed software and a continuous monitoring and incident resolution service for entire cement manufacturing lines.
The CemAI system has been installed at several cement manufacturing facilities in Europe, Africa, and North America.
“At CemAI, we assembled years of experience in cement manufacturing, inspection, and maintenance and then started to apply an AI and Machine Learning (ML) overlay to that expertise,” says Scott Ziegler, CEO of CemAI.
Taking a new approach
With a changing workforce of new employees replacing years of institutional knowledge, cement manufacturers needed a digital approach to replace and preserve their legacy products. Fortunately, newer employees are tech savvy, eager to learn AI, and realise how it can improve work life and elevate their job responsibilities. The ‘cement plant of the future’ exists now through human interaction with AI and ML.
CemAI manufacturing and maintenance experts teamed-up with AI technology company, Precognize (a Samson company), to test the combination of algorithms and sensor data at several pilot plants. Terabytes of historical plant process data needed to be managed and stored – a unique opportunity to digitalise the information. The team recognised that their process could be applied across the entire cement industry and CemAI was born. Current users are gaining impressive results using the technology for operating and managing ‘smart factories’.
Customers are provided with a ‘digital twin’ from the plant’s historical operation data as established by the existing sensors. This ‘twin’ creates a point of reference to detect even the smallest, seemingly insignificant deviations from standard operations. CemAI’s initial processes simulate the physical asset with a computer model and train it with any historical data available; three to six months of plant-wide data is ideal. A digital audit can identify gaps in plant monitoring. The CemAI system can be up and running in 8 – 12 weeks by their team of engineers with minimal plant involvement.
CemAI’s ‘specific-to-cement’ manufacturing algorithms are enhanced by the company’s remote monitoring centres. These centres, strategically located for full global service, receive alerts from the software in real-time. The alerts are analysed by cement plant maintenance experts to develop recommendations and remediation actions shared directly with plant teams. The remote monitoring centres house extensive libraries of prescriptive actions and report key performance indicators to each facility being monitored.
Enjoyed what you've read so far? Finish the article and access the entire December issue of World Cement by registering, free of charge, here.
Read the article online at: https://www.worldcement.com/special-reports/21122022/keeping-kilns-turning-with-machine-learning/
You might also like
CIMPOR celebrates 50 years of the Souselas Plant
The Plant of Souselas projects a sustainable future with a focus on decarbonisation and technological innovation.