HUBER+SUHNER accelerates data analysis from weeks to minutes with Azure Data Explorer

Headquartered in Switzerland, HUBER+SUHNER is a global company which develops and manufactures the essential building blocks for systems that require electrical or optical connectivity. With approximately 4,500 employees across 40 locations worldwide, it specializes in radio frequency, fiber optics, and low frequency technologies. "The components we produce are often unseen but critical. They are in planes, cars, trains, and charging stations," states Sascha Spitzhofer, Head IT Project and Service Delivery at HUBER+SUHNER.

The company’s production data—i.e. information about the manufacturing process that can include things like materials used, quantities produced, or issues—was scattered across multiple local databases. This made it cumbersome for HUBER+SUHNER’s data analysts to perform efficient, comprehensive data analysis required for traceability. “If there's an incident, we need to determine if our component was involved. We need to trace back to see if there were changes in production or materials. This helps identify whether other antennas may have the same issue,” explains Mauro Gerber, Data Scientist at HUBER+SUHNER.

"The trigger for our data transformation stemmed from our radar antennas for our automotive customers, which form part of the self-driving and assistance features in cars. We wanted to give our engineers insights into data so they could make real-time adjustments," recalls Gerber.  Another issue was scalability. In the past, the company only sent data to the database, and it was primarily used for source-specific purposes. However, as Gerber explains, "We also wanted the ability to access data, download data, and perform various operations on it. This two-way traffic became a bit of a problem because the SQL server couldn't handle simultaneous data reading and writing. If we attempted to read too much data or made an error in our data retrieval process, it could potentially crash the SQL server, bringing the entire production site to a halt."

Bridging the data divide

When it came to selecting the right data architecture, HUBER+SUHNER took a calculated approach. "Microsoft is a strategic partner for us, making it an easy decision to continue with them," said Spitzhofer. However, at the time, HUBER+SUHNER didn't have the engineering resources to implement the project in-house. “We started looking for experienced partners and found Option 4.0. Having them involved in the project afforded us the time to gradually upskill our own team,” adds Spitzhofer.

HUBER+SUHNER initially used Azure IoT Hub and Azure Edge to handle interactions and data processing for their Internet-of-Things (IoT) devices. They soon realized they needed more advanced databases to effectively manage the large volumes of data generated by these devices. To solve this, HUBER+SUHNER switched to using Azure Data Explorer (ADX) to enhance its data retrieval and analysis.

The company set up a new server cluster—i.e. a group of servers that work together, combining their capabilities for greater processing power, redundancy, and scalability. “Now, we aggregate process data from all plants into Azure. This approach allows for virtually infinite data storage, significantly enhancing scalability,” explains Luis Delgado, Chief Technical Officer at Option 4.0. The company made sure that creating data didn't slow down individual computers by spreading out data tasks among many machines or servers. "Each new entry in the local database triggers the IoT gateway to transfer the data to the ADX cluster, where we conduct our analysis," explains Michael Müssle, Team Lead Smart Factory at HUBER+SUHNER.

HUBER+SUHNER’s centralized data became the single source of truth for its global team. "Engineers in various locations can access the same data, minimizing errors," Gerber points out. This not only streamlined accessibility but also set the stage for advanced analytics. Müssle elaborates, “We connected ADX data to different analytics tools and created an overall equipment effectiveness (OEE) dashboard. This lets our engineers see the big picture and understand how materials, machinery, and settings impact the results. We've also created an app using Power Apps, with which engineers can scan a QR code on an antenna and trace its origin—like when and where it was produced. This fulfills the need for traceability and accountability.”

Transforming production through data

The benefits that HUBER+SUHNER has gained from the new data architecture are both substantial and measurable. "The new system built on ADX enables us to analyze data in minutes, not weeks. This speed enables engineers and quality teams to address problems immediately," Gerber emphasizes. Quality control teams can be more proactive rather than reactive, elevating the company's product standards. The company is already planning to extend this solution to 30 more machines next year. “We aim to collect detailed data for each item as we produce it and create a 'digital twin' for each one, helping us simulate and optimize its real-world performance, and ensuring the best quality and efficiency,” adds Müssle.

With this move, HUBER+SUHNER was also able to enhance its security posture. "Secure data is crucial for us. Using Azure services, we've modernized our approach," Müssle asserts. “There's no direct connection from the machines to the cloud. We use field gateways—or intermediate devices—to collect and publish data to the cloud.” With modernized IP addresses and enhanced firewalls, HUBER+SUHNER has strengthened its security protocols to keep its valuable data safe from unauthorized access.

Despite adopting a new database architecture, the company’s data scientists and analysts can continue using their current tools. “There's no need to overhaul existing systems due to the system's compatibility with our SQL servers. And the daily workflows and tasks of process engineers are uninterrupted, they can access data without disrupting the manufacturing line," adds Gerber.

Additionally, the new architecture has paved the way for more collaboration across departments. "The success of ADX has led to increased demand from other departments to integrate their data," Spitzhofer mentions. "This prevents us from creating data silos as was the case in the past. Instead of a project-based approach, we have a scalable template. Now, our engineering team no longer must wait months for new solutions, we can deploy new devices and apps in a couple of weeks.”

Scaling data impact

After successfully consolidating its data architecture, HUBER+SUHNER has a roadmap that extends beyond internal efficiency. The company aims to provide its customers with secure access to pertinent data, effectively turning data into a value-added service. This extends to scenarios where customers could optimize their products based on HUBER+SUHNER's real-time data. Gerber elaborates, "The goal is to connect the entire chain, from suppliers to customers, equipping both our team and our customers’ engineers and production staff to fine-tune devices for better performance, safety, and make informed decisions based on data."