Just as global corporations manage a heterogeneous work force spread across great distances, they also have to negotiate disparate engineering and enterprise systems.
Consider what Parker Hannifin faces. The diversified manufacturer is spread into 135 divisions, eight of which are in aerospace alone, according to Bob Deragisch, engineering services manager for the aerospace group. Those eight divisions use six different ERP systems and 185 disparate engineering tools.
In other words, creating a consistent, repeatable workflow is a mind-boggling challenge. While Deragisch emphasizes there is no single formula to address the litany of interoperability issues that global manufacturers are dealing with today, one major step he instituted is adopting a new suite of interoperability software as a means of ensuring data quality.
Data quality has long been an issue for global manufacturers. Assemblies, as well as parts, will often lose bits of information during each translation. While many Tier 1 suppliers might share the same CAD software, many more of its customers might not share the same release, causing slight variations in a design, according to Deragisch.
"It allows our engineers to design the original model as required with all the functionality of our own CAD system," says Deragisch. "Then, before we send our geometry to customers or suppliers, we employ CADFeature to get the data to the point where they can best utilize it if their system is at a lower revision than our standard."
Translation isn't the only issue manufacturers face. It's merely the first step. More manufacturers are requiring independent validation that the translation of their designs from one program (or version) to another is accurate.
Consider how the role of a supplier has changed. For decades, a supplier might have provided its products to just one company. But as supply chain diversity has increased, a Tier 2 company might be supplying dozens of customers, which means providing multiple CAD support, according to Ken Tashiro, Elysium's chief operating officer.
"If you're providing a machined part, you're relying on automated software processes like 3-D analysis and prototyping that minimize your time to deliver a product," says Tashiro. "But in order to do that, you need perfect data.
Parker Hannifin will use CADFeature to translate a design saved in one version of its Catia software and have it automatically validate the process.
"Each version of the CAD tool may use different mathematical representations to define geometry, and they have to be exact within a certain tolerance," says Deragisch. "I look at that tolerance as being our manufacturing tolerance. So the CAD translation needs to be within those decimal points of accuracy. There is very little room for error."
At a time when multiple systems, versions and divisions can create an interoperability mess for far-flung operations, automating the conversion process can eliminate human error, applying rules for converting features and mating constraints.