How Azure AI Platform Transforms Manufacturing: Practical Applications and Case Studies
Table of Contents

How Azure AI Platform Transforms Manufacturing: Practical Applications and Case Studies

Manufacturers are increasingly adopting digital technologies, and Microsoft plays a key role in this transformation through its cloud solutions. Companies are turning to cloud platforms because they offer more than just better performance—they help reduce costs and give businesses the flexibility to scale operations as needed. Cloud solutions also make it easier to manage large volumes of data and share information across teams and locations. Ultimately, these technologies allow manufacturers to streamline processes, improve decision-making, and focus more on innovation and long-term growth. By using cloud-based tools, they can build systems that support everything from day-to-day operations to more strategic initiatives like research and development.

This article examines how Microsoft’s Azure platform is being applied in real-world manufacturing settings. Frou case studies demonstrate the practical benefits: ASMPT’s improved production accuracy, Siemens’ advancements in predictive maintenance, Tikkurila's updates in using IoT and predictive maintanence and Epiroc’s operational optimizations.

Production Accuracy at ASMPT with Microsoft Azure Translator

ASM Pacific Technology (ASMPT), a leader in the semiconductor and electronics industries, faced the challenge of ensuring seamless communication across its global operations, particularly in a traditionally cloud-averse sector. By evolving its Factory Chat app into a web-based platform powered by Microsoft Azure, ASMPT improved production accuracy and efficiency and also enhanced collaboration across international teams.

Company background

ASM Pacific Technology (ASMPT) is a world market leader and technology pioneer in the semiconductor and electronics industries. Headquartered in Singapore, ASMPT supports manufacturers around the globe in establishing integrated smart factories. With a strong commitment to innovation and quality, the company has solidified its position as a key player in the industry. In 2023, ASMPT reported a revenue of US$1.88 billion, reflecting its significant influence and reach in the global market.

Challenge

ASMPT faced the challenge of ensuring effective knowledge transfer across its global locations in an industry that is traditionally cloud-averse. 

The semiconductor industry, which underpins sectors such as automotive, consumer electronics, data centers, LED lighting, and medical technology, is experiencing continuous growth in demand. To meet this demand, electronics manufacturers must constantly optimize their production processes. However, the ideal infrastructure varies from company to company, with complex production lines that require seamless integration of numerous machines.

Given the complexity, any disruption in machine operation can halt the entire production line, leading to significant economic losses and, in the case of medical devices, potential risks to human lives. To address these challenges, ASMPT initially introduced Factory Chat, a shop-floor communication app based on Microsoft Azure. 

However, as the Internet of Things (IoT) and Industrial Internet of Things (IIoT) became more integral to operations, the limitations of the on-premises solution became apparent. The existing infrastructure was insufficient for the scalability and availability needed for modern IoT applications, and there was significant industry resistance to adopting cloud-based solutions due to concerns over data security.

Solution

ASMPT responded by evolving Factory Chat into a completely web-based application that leverages Microsoft Azure's capabilities. 

Key elements of the solution include:

  • Web-Based Access: Factory Chat is now accessible to all of a company’s technicians worldwide, functioning like a social platform where machine operators can view specific problems and solutions from other sites.
  • Real-Time Translation: Azure Translator automatically and instantly translates all information into the local language, enabling seamless global communication.
  • Comprehensive Problem-Solving: Operators can manually enter problems or access a troubleshooting database, with detailed process definitions through shift reports, information posts, or audit data.
  • Machine-to-Human Communication: Machines can report errors directly to Factory Chat and autonomously assist in resolving issues.
  • Scalable Infrastructure: Factory Chat runs on Azure Kubernetes Services (AKS), allowing performance to be scaled according to workload and size, with simplified monitoring.
“By deploying Factory Chat on Azure, we’ve taken an important step toward the cloud and created a solid foundation for the future that we’re already benefiting from now—a kind of springboard for digital transformation.”  - Olimpiu-Petru Datcu: Project Lead, ASMPT

Impact

The implementation of Microsoft Azure Translator and the evolution of Factory Chat had a positive impact on ASMPT’s operations:

  • Improved Production Accuracy: Improve understanding of technical instructions across different languages, reduce errors and improve product quality.
  • Increased Efficiency: Streamlined communication and problem-solving led to greater overall efficiency in manufacturing processes.
  • Elevate Collaboration: Smoother collaboration across international teams resulted in more consistent and reliable operations.
  • Cost-Effectiveness: The move to Azure proved to be a cost-effective solution, boosting the overall performance and reliability of ASMPT’s systems.
“The more concentrated data we have, the more knowledge we have on performance and optimization. The IoT, and the IIoT in particular, are helping to make processes within factories more efficient.” - Olimpiu-Petru Datcu: Project Lead, ASMPT
“ With the help of Microsoft Azure, ASMPT has been able to design a complete high-performance system: from the application to hosting to authorization management. We’ve established an excellent platform that will allow us to build on that knowledge base going forward and add more applications.”  - Michael Rominger: Head of Business Development IIoT, ASMPT

The ASMPT's case study is available on the Microsoft website, and you can access it by clicking this link: https://customers.microsoft.com/en-us/story/1565004142676255055-asmpt-elunic-azure-translator-en

Transforming Predictive Maintenance at Siemens with Microsoft Azure

Siemens, a global leader in industrial automation and digitalization, revolutionized its maintenance operations by adopting AI-driven predictive maintenance models through Microsoft Azure. This transformation aimed to standardize AI development processes across the company’s global operations, significantly enhancing efficiency and reducing equipment downtime.

Background

Siemens is a global powerhouse in industrial automation and digitalization. With operations in over 200 countries, Siemens leads in engineering and technology, providing cutting-edge solutions for manufacturing, energy, healthcare, and infrastructure sectors. In 2023, Siemens reported revenues of €84.7 billion, reflecting its significant impact on the global market. The company employed approximately 303,000 people worldwide in 2021, further highlighting its vast operational reach and influence.

Challenge

Siemens faced the challenge of establishing a standardized process for developing AI models to optimize maintenance processes for industrial equipment across its global operations. AI-driven digital solutions offered tremendous potential to make processes more efficient, from reading metadata in PDF documents to detecting product damage through automated image analysis. 

However, the development of these AI solutions was hindered by a lack of standardized tools, code, and processes, leading to inefficiencies.

Previously, when departments had ideas for AI solutions to speed up business processes, they faced challenges in communicating these ideas to data analysts who lacked user-friendly tools to implement them. 

There was no common language or reusable components, which meant that each new AI model required starting from scratch, often taking several months to develop. This slow turnaround was a significant obstacle in delivering timely, impactful solutions. 

Siemens IT recognized the need for a platform that would provide standardized AI services, enabling faster development and broader participation from domain experts across the company. In response, Siemens launched the AI initiative in 2021, a platform fully based on Microsoft Azure, designed to streamline the development of AI models.

Solution

To address this challenge, Siemens implemented Microsoft Azure Machine Learning to develop and deploy advanced predictive maintenance models. Key elements of the solution include:

  • Data Integration: Siemens integrated data from sensors and industrial equipment across its operations into the Azure cloud platform, creating a centralized repository for analysis.
  • Machine Learning Models: Using Azure Machine Learning, Siemens developed sophisticated models that analyze historical and real-time data to predict equipment failures before they occur.
  • Scalability and Flexibility: Azure's cloud infrastructure allowed Siemens to scale its predictive maintenance solution globally, applying the models to a wide range of equipment and industries.
  • Continuous Improvement: The solution also enabled continuous refinement of the predictive models, using new data to improve accuracy and reliability over time.

Impact

The implementation of Microsoft Azure Machine Learning had a profound impact on Siemens’ maintenance operations:

  • Reduced Downtime: By predicting equipment failures before they happen, Siemens significantly reduced unplanned downtime, improving overall operational efficiency.
  • Cost Savings: The shift from reactive to predictive maintenance led to substantial cost savings by minimizing the need for emergency repairs and extending the lifespan of equipment.
  • Increased Reliability: The predictive maintenance models improved the reliability of Siemens’ equipment, leading to greater customer satisfaction and trust in their solutions.
  • Global Scalability: Siemens was able to deploy the solution across its global operations, standardizing maintenance practices and improving performance across multiple industries.
“We can now deploy compute clusters in Azure in just a few minutes. That wasn’t possible before.” - Dr. Ioannis Petrakis: IT Principal Key Expert Data Analytics & AI, Siemens IT

The Simens's case study is available on the Microsoft website, and you can access it by clicking this link: https://customers.microsoft.com/en-us/story/1637783244393505156-siemens-azure-machine-learning-en

Tikkurila: IoT Predictive Maintenance

Tikkurila, a renowned Nordic paint company, faced the challenge of modernizing its production line maintenance to reduce downtime, optimize costs, and improve product quality. By leveraging IoT for predictive maintenance, Tikkurila aimed to enhance operational efficiency, reduce product returns, and minimize warranty claims.

Background

Tikkurila, a prominent Nordic paint company with a rich history since 1862, develops premium surface products and services. Operating in 11 countries with over 2,400 employees, Tikkurila became part of PPG in June 2021, reporting EUR 582 million in revenue in 2020.

Challenge

Tikkurila faced significant challenges in optimizing production line maintenance. The company needed to minimize production downtime and reduce the rate of product returns and warranty claims. 

Traditional maintenance approaches were reactive and insufficient for the demands of modern, high-volume manufacturing. The goal was to implement an IoT-based predictive maintenance solution that could monitor equipment in real-time and predict maintenance needs, preventing costly disruptions and improving the overall production process.

Solution

Tikkurila deployed an IoT-based predictive maintenance system that utilized sensor data, including temperature, composition, environmental parameters, vibration, electric current, and voltage readings. 

These sensors, previously used only by the production line vendors, were integrated into Tikkurila’s global BI system. Predictive maintenance models based on neural networks, decision trees, and AFT models were employed to forecast equipment failures and optimize maintenance schedules.

The full cloud solution architecture ensured seamless integration with Tikkurila's business data analytics systems, allowing for real-time monitoring and adjustment of manufacturing parameters. 

By combining sensor data with predictive algorithms, Tikkurila was able to address equipment maintenance needs before significant disruptions occurred.

Impact

The implementation of the IoT predictive maintenance solution resulted in reduced production downtime, minimized product warranty claims, and improved overall product quality. 

The solution enabled better maintenance scheduling and workload management, ensuring equipment was maintained proactively rather than reactively. Additionally, Tikkurila was able to optimize manufacturing parameters and identify potential quality issues related to insufficient ingredients, further enhancing the quality of their products.

Interested in Learning More? If you'd like to learn more about our collaboration with Tikkurila, you can check the client story here: https://www.itmagination.com/clients/tikkurila

Optimizing Operations at Epiroc with Microsoft Azure 

Epiroc, a leader in mining and infrastructure equipment, sought to enhance its manufacturing and operational processes to meet the industry's growing complexity and demands. By implementing Microsoft Azure's suite of tools, including Azure Machine Learning and Azure Databricks, Epiroc developed a robust, scalable data platform that enabled predictive maintenance, improved product quality, and reduced waste. 

Background

Epiroc is a leading productivity partner for the mining, infrastructure, and natural resources industries. Headquartered in Stockholm, Sweden, Epiroc develops and produces innovative equipment, machinery, and tools that improve productivity, safety, and sustainability. 

The company operates in more than 150 countries, delivering essential solutions for the modern mining and infrastructure sectors. In 2023, Epiroc reported revenues of over SEK 60 billion, supported by approximately 18,200 passionate employees who collaborate closely with customers worldwide.

Challenge

Epiroc faced the challenge of optimizing its manufacturing and operational processes to meet the growing demands of its customers in the mining and infrastructure industries. 

As equipment and machinery became more advanced, the complexity of operations increased, leading to challenges in maintaining efficiency and minimizing downtime. 

Epiroc identified the opportunity to leverage data and machine learning to gain deeper insights into their operations, boost predictive maintenance, and optimize production processes. 

However, the company needed a robust and scalable platform to handle the vast amounts of data generated and to develop accurate predictive models that could be deployed across its global operations.

Solution

To address this challenge, Epiroc implemented a comprehensive solution using Microsoft Azure Machine Learning, Azure Data Factory, and Azure Databricks. Key aspects of the solution include:

  • Data Heaven Creation: Leveraging Microsoft Azure’s robust ecosystem, Epiroc created a "data heaven" where data from across the organization was integrated and made accessible for analysis and machine learning. This included the use of Azure Data Factory for data integration and Azure Databricks for advanced data processing.
  • Sustainability Alignment: Microsoft’s existing policies and processes aligned with Epiroc’s sustainability objectives, providing confidence that the solution would not only meet key performance indicators but also support ongoing improvements in sustainability.
  • AI Factory Implementation: With guidance from the Microsoft Customer Success Team in Sweden, Epiroc set up an AI factory using the Enterprise Scale Machine Learning (ESML) framework. This AI factory, built on Azure, incorporated services such as Azure Machine Learning, Power BI, and a data lake architecture supporting data mesh.
  • Rapid Deployment: In collaboration with Microsoft Cloud, Epiroc established the AI factory in just 60 hours, ensuring a secure and scalable environment with private networking for deploying predictive maintenance and operational optimization models.
Image source: https://customers.microsoft.com/en-us/story/1653030140221000726-epiroc-manufacturing-azure-machine-learning 

Impact

The implementation of Microsoft Azure Machine Learning had a significant impact on Epiroc’s operations:

  • Improved Steel Quality and Waste Reduction: The new solution enabled Epiroc to achieve consistent steel quality, leading to a reduction in waste. Predictive AI helped the company consistently produce higher-quality material, reducing customer rejections and product returns by 30 percent, which saved time and money.
  • Cost Savings from Logistics: Higher product quality also resulted in fewer trips to retrieve unwanted deliveries, leading to additional cost savings.
  • Accelerated Innovation: The rapid implementation of the solution inspired Epiroc to explore further innovations. The company now has eleven analytical teams with active use cases in the ESML AI Factory, with plans for more in the coming year.
  • Competitive Edge: The ability to analyze extensive data from production processes, as well as research and design, has positioned Epiroc to outperform the competition, accelerate innovation, and adapt quickly to customer needs.
  • Global Standardization: The scalable nature of Azure allowed Epiroc to standardize its operational processes across all global sites, ensuring consistency and reliability in production.
“Now, with Azure Machine Learning and predictive AI we consistently produce material with better quality.” - Peter Malmberg: Vice President Digitalization

The Epiroc's case study is available on the Microsoft website, and you can access it by clicking this link: https://customers.microsoft.com/en-us/story/1653030140221000726-epiroc-manufacturing-azure-machine-learning

Conclusion

At ASMPT, the transition to a cloud-based communication platform using Azure Translator has significantly improved production accuracy and collaboration across global teams. Siemens has transformed its maintenance processes with predictive AI models, reducing downtime and increasing equipment reliability on a global scale. Tikkurila has modernized its production line with IoT-based predictive maintenance, reducing product returns and warranty claims while enhancing product quality and operational efficiency. Epiroc’s integration of Azure Machine Learning and AI has not only optimized their operations but also driven innovation and sustainability within their manufacturing processes.

For more information on how Microsoft Azure's cloud solutions can reshape your business, contact us and let’s discuss how we can support you.

Liked the article? subscribe to updates!
360° IT Check is a weekly publication where we bring you the latest and greatest in the world of tech. We cover topics like emerging technologies & frameworks, news about innovative startups, and other topics which affect the world of tech directly or indirectly.

Like what you’re reading? Make sure to subscribe to our weekly newsletter!
Relevant Expertise:
Share

Join 17,850 tech enthusiasts for your weekly dose of tech news

By filling in the above fields and clicking “Subscribe”, you agree to the processing by ITMAGINATION of your personal data contained in the above form for the purposes of sending you messages in the form of newsletter subscription, in accordance with our Privacy Policy.
Thank you! Your submission has been received!
We will send you at most one email per week with our latest tech news and insights.

In the meantime, feel free to explore this page or our Resources page for eBooks, technical guides, GitHub Demos, and more!
Oops! Something went wrong while submitting the form.

Related articles

Our Partners & Certifications
Microsoft Gold Partner Certification 2021 for ITMAGINATION
ITMAGINATION Google Cloud Partner
AWS Partner Network ITMAGINATION
ISO-IEC 27001:2013 ITMAGINATION ISO 9001 ITMAGINATION
© 2024 ITMAGINATION, A Virtusa Company. All Rights Reserved.