Demand a standard platform for data collection
By Dave Perkon, technical editor for Control Design
The cost of recent automotive recalls and downtime reinforces the need for improved quality management and parts traceability across the industry’s complex, global supply chains. As a supplier of driveline technologies to some 90% of the world’s car manufacturers, GKN Driveline needed to up its traceability game, said Geoffrey Bennett, senior enterprise applications business analyst for the $12 billion designer, manufacturer and servicer of OEM systems and components.
There is a long history of vertical integration, mergers and acquisitions that required GKN to bring together many different traceability systems, some relying on vulnerable processes based on spreadsheets and paper-based records. At Smart Industry 2017, Bennett discussed how a new, scalable approach is bringing integration and uniformity to the company’s dissimilar systems, allowing 100% visibility into parts history “with a couple of mouse clicks.”
Headquartered in Redditch, United Kingdom, GKN Driveline is a rather large—with 26,000 employees and 46 manufacturing sites in 23 countries—and very old organization, at 250 years and counting. "We have a wide footprint and because of that we need solutions that can be easily scaled to the enterprise level," said Bennett. “They must also be easy to maintain because we have a very large, complex business and supply chain. It needs to just work."
For its driveline division, most of GKN Driveline's manufacturing is in North Carolina, with three main plants in the drive-shaft business. At the Sanford plant, the operation starts with raw steel and forges it into a shape; at the Alamance plant, parts are machined into shapes; and the Roxboro facility final-assembles the
components. This final assembly process is where GKN Driveline focused its traceability efforts, as it ships products directly to automotive manufacturers.
A range of legacy systems
"At each plant we had requirements to track processes and improve them, but each plant had very different process and competencies," said Bennett. "My first two years at GKN, I was in quality. The past three years, I've been in IT, and much of the work I do is relieving pain points I experienced when I was on the quality side of the house. If a part came back from a customer, I spent a great deal of time sifting through paper to find traceability information with many an afternoon spent digging though boxes trying to find something."
To lessen the use of paper, documents were scanned, automated data collection was implemented. But the methods were not uniform across the plants. "Plant A did it one way, Plant B did it differently, and so on," said Bennett. "Even in the same facility, in different departments, the automated data collection used different methods. We were automatically collecting the data, but the different systems were not talking to each other."
GKN's data wasn't aligned to where it was usable across the facilities. "Not having a unified standard presents a lot of challenges," said Bennett. "We needed to provide our customers with bar-to-car traceability. We all agreed on what was needed; the problem was how to do it."
Using the company’s ERP system was considered, as all the plants are connected to it, but there were various issues. "We looked for tools that were standardized," said Bennett. "We wanted to do things like remote diagnostics and configuration and provide easy remote access to the data. In the end, we selected Argonaut Manufacturing Performance Platform from Red Viking."
A key point was that Argonaut allowed GKN to link all of its plants together under a single platform, continued Bennett. "It also had flexibility and scalability to expand to a region or even worldwide."
Standardized applications were deployed to its manufacturing cells at the assembly level using edge-of-network IoT devices called spearheads. These devices can be controlled, managed and changed remotely and include all the protocols needed communicate to machine controllers on the plant floor, simplifying integration.
"Track-and-trace is the bread-and-butter application," said Bennett. "It is collecting data from each manufacturing operation. This includes critical data such as press, force and position, which is tied to the serial number of the part. We are collecting the data to create a comprehensive record."
It is no longer necessary to look through boxes in a warehouse or search through Google Docs to find needed information. "With just a serial number, what used to take hours to find now takes seconds," said Bennett. "In a similar way, the pack-out application allows us to send a very detailed part manifest to the customer. This detailed information allows customers to find all the parts in a shipment, helping to track down any problem parts."