Download PDF
Aluminerie Alouette implements STATISTICA Data Miner and MSPC
Technology Category
- Analytics & Modeling - Data Mining
- Analytics & Modeling - Machine Learning
- Analytics & Modeling - Predictive Analytics
Applicable Industries
- Metals
Applicable Functions
- Process Manufacturing
- Quality Assurance
Use Cases
- Machine Condition Monitoring
- Predictive Maintenance
- Process Control & Optimization
Services
- Software Design & Engineering Services
- Training
The Challenge
Aluminerie Alouette needed to continuously improve its production processes to stay among the worldwide leaders in aluminum manufacturing. The company faced the challenge of understanding the influence of several hundred inputs on the aluminum manufacturing output. Some inputs could be controlled, such as the dosage of additives and energy management, while others, like outside temperature and raw material composition, could not. To address this, Aluminerie Alouette required a solution that could identify significant inputs and develop multivariate models to monitor key performance indicators.
About The Customer
Aluminerie Alouette, established in 1992, is an independently operated company producing primary aluminum. With a workforce of 1,000 employees and an annual production capacity exceeding 600,000 tons, it stands as the largest employer in Sept-Îles, Canada, and the leading aluminum smelter in the Americas. The Sept-Îles smelter is renowned globally for its energy consumption efficiency and state-of-the-art technology, surpassing government environmental standards. Aluminerie Alouette has been utilizing STATISTICA Enterprise for several years to monitor key performance indicators and control the production process efficiently.
The Solution
To better understand the influence of various inputs on the aluminum manufacturing process, Aluminerie Alouette augmented its existing STATISTICA Enterprise with STATISTICA Data Miner and MSPC software for multivariate analyses. These StatSoft modules were expected to identify inputs with significant influence on key performance indicators and develop multivariate models for monitoring additional indicators. The implementation of these modules allowed Aluminerie Alouette to conduct several analyses, validating that STATISTICA Data Miner and MSPC met their needs. Automated neural networks in STATISTICA enabled the development of models representing different operating scenarios based on historical data. These models allowed for the adjustment of relevant inputs without compromising the quality of operations and facilitated medium-term production predictions considering future events.
Operational Impact
Quantitative Benefit
Related Case Studies.
Case Study
Goldcorp: Internet of Things Enables the Mine of the Future
Goldcorp is committed to responsible mining practices and maintaining maximum safety for its workers. At the same time, the firm is constantly exploring ways to improve the efficiency of its operations, extend the life of its assets, and control costs. Goldcorp needed technology that can maximize production efficiency by tracking all mining operations, keep employees safe with remote operations and monitoring of hazardous work areas and control production costs through better asset and site management.
Case Study
KSP Steel Decentralized Control Room
While on-site in Pavlodar, Kazakhstan, the DAQRI team of Business Development and Solutions Architecture personnel worked closely with KSP Steel’s production leadership to understand the steel production process, operational challenges, and worker pain points.
Case Study
Bluescope Steel on Path to Digitally Transform Operations and IT
Increasing competition and fluctuations in the construction market prompted BlueScope Steel to look toward digital transformation of its four businesses, including modern core applications and IT infrastructure. BlueScope needed to modernize its infrastructure and adopt new technologies to improve operations and supply chain efficiency while maintaining and updating an aging application portfolio.
Case Study
RobotStudio Case Study: Benteler Automobiltechnik
Benteler has a small pipe business area for which they produce fuel lines and coolant lines made of aluminum for Porsche and other car manufacturers. One of the problems in production was that when Benteler added new products, production had too much downtime.
Case Study
Continuous Casting Machines in a Steel Factory
With a very broad range of applications, steel is an important material and has been developed into the most extensive alloy in the engineering world. Since delivering high quality is absolutely crucial for steel plants, ensuring maximum productivity and the best quality production are the keys to competitiveness in the steel industry. Additionally, working conditions in steel factories are not suitable for workers to stay in for long periods of time, so manufactures usually adopt various machines to complete the steel production processes. However, the precision of these machines is often overestimated and the lack of flexibility also makes supervisors unable to adjust operating procedures. A renowned steel factory in Asia planned to improve its Distributed Control System (DCS) of furnaces as well as addressing the problem of insufficient accuracy. However, most well-known international equipment suppliers can not provide a satisfactory solution and local maintenance because the project needed new technologies to more accurately control equipment operations. By implementing Advantech’s automated monitoring and control solution, steel factories can not only improve the manufacturing processes but can also allow users to add additional functions to the existing system so as to make sure the operation runs at high efficiency.
Case Study
Automated Predicitive Analytics For Steel/Metals Industry
Asset to be monitored: Wire Compactor that produces Steel RebarCustomer Faced The Following Challenges:Dependent upon machine uptime.Pressure cylinders within the compactor fail to control compression and speed causing problems in binding the coil.Equipment failure occurs in the final stage of production causing the entire line to stop, can you say bottleneck?Critical asset unequipped with sensors to produce data.
