The U.S. automotive industry has been on a wild ride since 2008 and new light vehicle sales in the U.S. have been increasing year over year, setting a new sales record of 17.6 million vehicles in 2016. The global market is stable, which is good news for tier-one automotive suppliers like Hirotec Group Companies.
Hirotec manufactures car doors and vehicle exhaust systems in 26 facilities across the globe and counts General Motors and Mazda among its customers. The company designs and produces more than 7 million doors and 1.5 million exhaust systems a year. Last year, it began investigating how to move away from reactive maintenance to see how a cloud platform could help improve condition monitoring of computer numerical control (CNC) machines at its Detroit facility.
The Detroit facility relies on CNC-based machines to fabricate components for the company’s automotive doors and exhaust systems, where production consists of many real-time variables, such as feed rate, tool path, direction and the speed of servo motors.
To support its long-term Industrial Internet of Things (IIoT) vision, Hirotec collaborated with PTC on short, six-week Agile sprints. For the first Agile pilot project, PTC’s ThingWorx targeted legacy CNC-based machines and more modern equipment while Hirotec focused on the wide range of data types with these eight machines.
“With the first IIoT project, we wanted to prove to ourselves that we could collect data from all sorts of different vintage machines,” says Justin Hester, senior researcher in Hirotec’s IoT Lab. “We wanted to keep the scope small so that we could run it within that small sprint time.”
Hirotec turned to KEPServerEX and the ThingWorx IoT Platform to enable the transport of variable data types from the plant floor to cloud analytics. “What Hirotec had to do first was come up with a strategy to bring in the same data across all the machines,” Hester says. “What is the data we want to standardize and can reasonably collect from all our machines?”
Before the implementation of this pilot project, operations and maintenance data resided across multiple data silos and prevented access to actionable data on the plant floor. “A lack of data was never an issue for us,” Hester says. “As one of the largest automotive manufacturing suppliers in the world, we collect volumes of datasets on a daily basis. The problem we faced was transitioning from a data-heavy organization to a data-smart organization.”
For this tricky endeavor, Hirotec uses Kepware’s IoT Gateway for KepServerEX. This technology allows data points to move from CNC machines to third-party endpoints, and ultimately web-based browsers. The data can include item ID, value, quality and timestamp in a standard JavaScript Object Notation (JSON) format. Configurable headers are available for Representational State Transfer (REST) and Message Queue Telemetry Transport (MQTT) clients, TLS/SSL encryption on all agents, and data collection rates as frequent as 10 ms up to once per 27.77 hours for REST and MQTT clients, according to Kepware.
“For the modern CNC machines, we plug into Ethernet and connect through industrial protocols directly to that machine,” Hester says. “But in other cases, we had to install I/O blocks and tap into the hardwired controls of the machines, then transfer that machine data through Modbus TCP to KEPServerEX and transfer the data to the ThingWorx cloud.” The eight-machine testing ground at the plant has about 100 I/O points.
With the cloud-based platform in place, Hirotec provided a roadmap to ThingWorx for a range of data analytics and reporting options at different levels of the enterprise. At the management level, any web-based device can securely log into the system and see a high-level dashboard view of production, while a medium-level access reveals the status of the shop floor via local web-based displays.
“However, managers and team members working in that production area can log in straight from their laptops and iPhones,” Hester says. “There, they can get very detailed data and analysis in real time of the shop.”
With this cloud-based platform, IT assistance is minimized as operators and plant technicians are able to modify different operating metrics with clear, drag-and-drop interfaces. “If we want to see a line chart of this KPI, drag an icon over onto the screen,” Hester explains. “The system asks what data point do you want to show, choose it and the programming is done.”
Security is vital as data moves to the cloud, Hester adds. The Detroit Agile project relied on PTC’s cloud service to implement the instance of ThingWorx, which uses Amazon Web Services. Late last year, PTC received IoT Competency status—a milestone that recognizes “relevant technical proficiency and proven customer success.”
The wild ride for Hirotec and the automotive industry continues as this global company plans on more investments in these types of cloud solutions for its entire enterprise. More innovation to come…stay tuned.
